JP4207109B2 - Data conversion method, data conversion apparatus, data reproduction method, data restoration method, and program - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a data conversion method and data conversion device, a data reproduction method and data reproduction device, a data restoration method and data restoration device, a data format, a recording medium, and a program, and in particular, when content preview data is distributed to users. The present invention relates to a data conversion method and a data conversion device, a data reproduction method and a data reproduction device, a data restoration method and a data restoration device, a data format, a recording medium, and a program that are suitable for use in a computer.
[0002]
[Prior art]
For example, due to the spread of communication network technologies such as the Internet, improvement of information compression technology, and the progress of higher integration or higher density of information recording media, audio, still images, moving images, or audio For example, digital contents composed of various multimedia data such as movies made of moving images are distributed to viewers for a fee via a communication network.
[0003]
For example, a store that sells package media such as CD (Compact Disk) and MD (Mini-Disk) (trademark), that is, a recording medium on which digital contents are recorded in advance, for example, a large number of music data and the like. By installing an information terminal such as a so-called MMK (Multi Media KIOSK) in which digital content is stored, it is possible to sell not only package media but also digital content.
[0004]
The user inserts a recording medium such as MD brought into the MMK, refers to the menu screen, selects the title of the digital content to be purchased, and pays for the requested content. The payment method may be cash insertion, electronic money exchange, or electronic payment using a credit card or prepaid card. The MMK records the selected digital content data on a recording medium inserted by the user by a predetermined process.
[0005]
As described above, a seller of digital contents can distribute digital contents to users via the Internet, for example, in addition to selling digital contents to users using MMK.
[0006]
In this way, not only selling package media in which content is recorded in advance, but also by introducing a technique for selling digital content itself, content has been distributed more effectively.
[0007]
In order to distribute digital contents while protecting the copyright, for example, by using a technique such as Japanese Patent Application Laid-Open No. 2001-103447 or Japanese Patent Application Laid-Open No. 2001-325460, a part other than the portion where the digital content can be auditioned is encrypted. Only the user who has distributed and purchased the decryption key for encryption can be allowed to listen to all the contents. As an encryption method, for example, an initial value of a random number sequence that is a key signal for a bit string of PCM (Pulse Code Modulation) digital audio data is given, and the generated random number sequence of 0/1 and PCM data to be distributed are A method is known in which an exclusive OR is used as an encrypted bit string. The digital content encrypted in this manner is distributed to users by being recorded on a recording medium using the above-described MMK or distributed via a network, for example. If the user who has obtained the encrypted digital content data does not get the key, he or she can only listen to the unencrypted part that can be auditioned, and plays back without decrypting the encrypted part. Even so, only noise can be auditioned.
[0008]
In addition, technologies for compressing and broadcasting audio data and the like, distributing it via a network, and recording compressed data on various types of recording media such as magneto-optical disks have also been improved.
[0009]
There are various methods for high-efficiency encoding of audio data. For example, the audio signal on the time axis is not blocked, but is divided into a plurality of frequency bands and encoded (SBC (Sub) Band Coding)), or a block frequency band division method (so-called transform coding) that performs spectrum conversion of a signal on the time axis into a signal on the frequency axis, divides the signal into multiple frequency bands, and encodes each band. and so on. In addition, after band division by band division coding, a method is conceived in which, in each band, a signal is spectrum-converted into a signal on the frequency axis and coding is performed for each spectrum-converted band.
[0010]
  The filter used here is, for example, QMF (Quadrature Mirror Filter). Regarding QMF, “Digital coding of speech in subbands” (Bell Syst. Tech. J. Vol. 55, No. 8) by RE Crochiere. 1974). Also, “Polyphase Quadrature by Joseph H. RothweilerFiltersDocuments such as “-A new subband coding technique” (ICASSP 83, BOSTON) describe filter division techniques with equal bandwidth.
[0011]
As the above-described spectral transformation, for example, the input audio signal is blocked in a predetermined unit time (frame), and discrete Fourier transform (DFT), discrete cosine transform (DCT) is used for each block. Transform), modified DCT transform (MDCT; modified Discrete Cosine Transform), and the like. For example, details of MDCT can be found in the paper "Subband / Transform Cording Using Filter Bank Designs Based on Time Domain Aliasing Cancellation" (ICASSP 1987) by JP Princen, AB Bradley (Univ. Of Surrey Royal Melbourne Inst. Of Tech.) And others. It is stated in.
[0012]
Further, when the above-described DFT or DCT is used as a method for spectrally converting a waveform signal, M independent real data can be obtained by performing conversion in a time block composed of M samples. In order to reduce the connection distortion between time blocks, N / 2 samples are usually overlapped with each adjacent block, that is, N samples are overlapped on both sides, so in DFT and DCT, on average, , (M + N) samples of independent M pieces of real number data are quantized and encoded.
[0013]
On the other hand, when the above-described MDCT is used as a spectrum conversion method, when the conversion is performed with a time block composed of M samples, M / 2 each of the adjacent blocks, that is, both sides, Since M independent real data is obtained from 2M samples that are overlapped by M in total, in MDCT, on average, M real data is quantized for M samples. Will be encoded.
[0014]
In the decoding apparatus, a waveform signal can be reconstructed by adding waveform elements obtained by inversely transforming each block from codes obtained using MDCT while interfering with each other.
[0015]
In general, lengthening the time block for conversion increases the frequency resolution of the spectrum and concentrates energy on specific spectral components. Therefore, by using the MDCT which performs conversion with a long block length by overlapping with both adjacent blocks by half, and the number of obtained spectrum signals does not increase with respect to the number of time samples as a basis. By performing conversion, encoding can be performed more efficiently than when DFT or DCT is used for conversion. Further, by providing a sufficiently long overlap between adjacent blocks, it is possible to reduce distortion between blocks of the waveform signal.
[0016]
As described above, by quantizing the signal divided for each band by filtering or spectral conversion, it is possible to control the band in which the quantization noise is generated. Therefore, more efficient encoding can be performed. In addition, before performing quantization, by performing normalization for each band, for example, with the maximum absolute value of the signal component in the band, further efficient encoding can be performed.
[0017]
When quantizing each frequency component divided into frequency bands, for example, the frequency division width may be determined in consideration of human auditory characteristics. In other words, the audio signal may be divided into a plurality of bands (for example, 25 bands) so that the higher the band generally called a critical band (critical band), the wider the bandwidth.
[0018]
In addition, when the band is divided so that the critical band becomes wide, when data for each band is encoded, predetermined bit allocation may be performed for each band, or adaptation may be performed for each band. Alternatively, bits may be assigned (bit allocation is performed).
[0019]
For example, when coefficient data obtained by MDCT is encoded by bit allocation, the number of bits is adaptively allocated to MDCT coefficient data for each band obtained by MDCT for each block, Encoding is performed. As the bit allocation method, for example, the following two methods are known.
[0020]
R. Zelinski, P. Noll et al., "Adaptive Transform Coding of Speech Signals" (IEEE Transactions of Acoustics, Speech, and Signal Processing, Vol. ASSP-25, No. 4, August 1977) It is described that bit allocation is performed based on the signal size. According to this method, the quantization noise spectrum is flattened and the noise energy is minimized, but it is optimal in terms of reducing the noise that can actually be heard by the human ear because the masking effect is not used when considered auditorily. is not.
[0021]
In addition, the paper of “The critical band coder digital encoding of the perceptual requirements of the auditory system” (ICASSP 1980) by MA Kransner (Massachusetts Institute of Technology) is required for each band by using auditory masking. A technique for obtaining a stable signal-to-noise ratio and performing fixed bit allocation is described. However, in this method, even when the characteristic is measured with a sine wave input, the bit allocation is fixed, and the characteristic value is not so good.
[0022]
In order to solve these problems, all bits that can be used for bit allocation are divided into fixed bit allocation patterns determined in advance for each small block and bit allocation depending on the signal size of each block. High-efficiency coding in which the division ratio is dependent on the signal related to the input signal, and the division ratio into the fixed bit allocation pattern is increased as the spectrum of the signal is smoother. A device has been proposed.
[0023]
By using this method, when energy is concentrated in a specific spectrum, such as a sine wave input, a large number of bits can be assigned to the block including the spectrum, so the overall signal-to-noise characteristics Can be significantly improved. In general, human hearing for signals with steep spectral components is very sensitive, so improving signal-to-noise characteristics using such methods is not only a characteristic value for measurement, but also human characteristics. Is effective in improving the quality of the sound that you actually hear.
[0024]
Many methods other than those described above have been proposed for bit allocation. In addition, the auditory model has been refined and the encoding device has been improved so that not only characteristic values for measurement but also human hearing can be encoded more efficiently. Yes. In these methods, a real bit allocation reference value is obtained so that the signal-to-noise characteristic obtained by calculation is realized as faithfully as possible, and an integer value approximating it is obtained and set to the number of allocated bits. It is common to be done.
[0025]
Further, Japanese Patent Application No. Hei 5-152865 or WO94 / 28633 filed earlier by the present inventor discloses a tone component that is particularly important from the viewpoint of the generated spectrum signal, that is, around a specific frequency. A method is described in which a component in which energy is concentrated is separated and encoded separately from other spectral components. By this method, it is possible to effectively encode an audio signal or the like at a high compression rate with almost no auditory degradation.
[0026]
When generating an actual code string, first, for each band in which normalization and quantization are performed, quantization accuracy information and normalization coefficient information are encoded with a predetermined number of bits, then normalization, and The quantized spectral signal is encoded. Also, ISO / IEC 11172-3; (1993 (E), a933) describes a high-efficiency encoding method that is set so that the number of bits representing quantization accuracy information differs depending on the band. It is standardized so that the number of bits representing quantization accuracy information decreases as it becomes a region.
[0027]
Instead of directly encoding quantization accuracy information, for example, a method of determining quantization accuracy information from normalized coefficient information in a decoding device is also known, but in this method, when the standard is set, Since the relationship between the quantization coefficient information and the quantization accuracy information is determined, it becomes impossible to introduce control using quantization accuracy based on a more advanced auditory model in the future. In addition, when there is a range in the compression rate to be realized, it becomes necessary to define the relationship between the normalization coefficient information and the quantization accuracy information for each compression rate.
[0028]
As a method for encoding quantized spectral signals more efficiently, for example, it is described in a paper of “A Method for Construction of Minimum Redundancy Codes” (Proc. IRE, 40, p. 1098, 1952) by DA Huffman. There is also known a method for efficiently performing encoding using a variable length code.
[0029]
The signal encoded by the method described above can be encrypted and distributed in the same manner as the PCM signal. When this scrambling method is used, the key signal is obtained. Those that do not can reproduce the original signal. In addition, there is a method of encoding for compression after converting a PCM signal into a random signal instead of encrypting the encoded bit string. However, when this scrambling method is used, the key signal is Those that are not available can only reproduce noise.
[0030]
Further, distribution of content data trial data can promote sales of content data. The sample data includes, for example, data reproduced with lower sound quality than the original data, data capable of reproducing a part of the original data (for example, only the rust portion), and the like. When the user plays the trial listening data and likes it, he / she can purchase a key to decrypt the encryption so that the original sound can be played back, or a new recording medium on which the original sound data is recorded can be obtained. Try to buy.
[0031]
However, in the above-described scramble method, the entire data cannot be reproduced or all is reproduced as noise. For example, a recording medium in which sound is recorded with a relatively low sound quality is used for distribution as audition data. I could not do it. Even if the data scrambled by these methods is distributed to the user, the user cannot grasp the outline of the entire data.
[0032]
In addition, in the conventional method, when encrypting a signal that has been subjected to high-efficiency encoding, the compression efficiency is not lowered while giving a meaningful code string to a reproduction apparatus that is usually widely used. It was very difficult to do. That is, as described above, when a code string generated by performing high-efficiency encoding is scrambled, even if the code string is reproduced without being descrambled, not only noise is generated, but also by scrambling. If the generated code string does not conform to the original high efficiency code standard, there is a possibility that the reproduction process cannot be executed at all.
[0033]
Conversely, when high-efficiency encoding is performed after scrambling a PCM signal, for example, if the amount of information is reduced using the auditory property, irreversible encoding is performed. Therefore, even when such a high-efficiency code is decoded, a signal obtained by scrambling a PCM signal cannot be reproduced correctly. That is, such a signal is very difficult to correctly descramble.
[0034]
Therefore, a method has been selected that can correctly scramble even if the compression efficiency decreases.
[0035]
In response to such a problem, the present inventors, for example, disclosed in Japanese Patent Laid-Open No. 10-135944, only the code corresponding to the high band was encoded out of the music data converted into a spectrum signal and encoded. Disclosed is an audio encoding system that enables even a user who does not have a key to decrypt and reproduce an unencrypted narrowband signal by distributing the data as trial listening data. In this method, the high frequency side code is encrypted, the high frequency bit allocation information is replaced with dummy data, and the high frequency true bit allocation information is reproduced by the decoder that performs the reproduction process. Information is recorded at a position where it is not read (ignored) during processing.
[0036]
By adopting this method, the user receives distribution of the audition data, reproduces the audition data, purchases a key for decrypting the favorite audition data into the original data as a result of the audition, and pays for it. It will be possible to play music of all types correctly and enjoy it with high sound quality.
[0037]
[Problems to be solved by the invention]
In the technique disclosed in Japanese Patent Laid-Open No. 10-135944, a user who does not have a key can decrypt only a narrowband signal of data distributed free of charge. However, since its security depends only on encryption, if the encryption is decrypted, the user can play high-quality music without paying a fee. A data distributor (content provider) cannot legitimately collect charges.
[0038]
In addition, the content provider does not limit the quality of the entire content of the audition data, but enables the audition by limiting the quality of only a part of the content or only a few places, and the quality of the other parts Even if the data is restricted, there are cases where it is desired that the audition cannot be performed.
[0039]
For example, when distributing audition data free of charge, if you want to allow the user to audition only by limiting the quality of the rusted part of the song to tens of seconds, It is desirable to make it impossible to listen to a part other than the part that can be auditioned, and to create trial data that can reproduce the part that can be auditioned unnaturally.
[0040]
As described above, by using a technique that prevents a part that cannot be auditioned from being auditioned, even if the data is reproduced, the corresponding part (part other than rust) can be silenced. However, when listening to sample data that is silent except for the rust portion and only the rust portion is reproduced with low quality, the decoding process is performed sequentially from the beginning of the data, so that the portion that can be auditioned Is reproduced without sound, and it is unnatural that no sound is reproduced and output until the part of the data that can be auditioned is decoded.
[0041]
The present invention has been made in view of such a situation, and when distributing audition data in which only one part of the content data can be reproduced with low sound quality and the other part cannot be reproduced, only the reproducible part is distributed. Can be reproduced naturally, and the audition data in which the header length of the audition data is not changed can be generated, and such audition data can be restored to the original data.
[0042]
[Means for Solving the Problems]
  The data conversion method of the present invention is included in the first data string., At least part of the data used to reproduce the predetermined informationThe first data, When used for reproducing the predetermined information, the reproduction quality of the predetermined information is deteriorated as compared with the case of the first data string.The first replacement step for replacement with the second data and the first data are included in different areas., At least part of the data used to reproduce the predetermined informationThird dataAt least some ofThe, Control information referred to when reproducing the second data stringReplace with 4th dataAt the same time, the other part is a non-reference area that is not referred to as reproduction data when the second data string is reproduced.A second replacement step, a first replacement step, and a first generation step for generating a second data string using the data generated by the second replacement step and the data generated by the second replacement step.IncludeIt is characterized by that.
[0044]
  The first data string and the second data string can each be composed of a plurality of frames, and the third data can be recorded in a plurality of frames,4th dataCan be control information indicating whether or not the corresponding frame can be reproduced.
[0045]
  A second generation step of generating a third data string necessary for restoring the second data string generated by the processing of the first generation step into the first data string; The third data string generated by the process of the second generation step can beA portion that includes the first data and is replaced with the fourth data by the processing of the second replacement step in the third dataCan be included.
[0054]
  A frequency component conversion step for converting the input data into frequency components, and an encoding step for encoding the data converted into frequency components by the processing of the frequency component conversion step can be further included. In the process of 1 replacement step, the encoded data encoded by the process of the encoding step can be used as the first data string, and the first data can be replaced with the second data.In the process of the second replacement step, the encoded data encoded by the process of the encoding step is used as the first data string, and at least a part of the third data is replaced with the second data. can do.
The third data includes spectral coefficient information of the frequency component converted by the processing of the frequency component conversion step.At least some ofCan be included.
The first data may include at least a part of the normalization coefficient information of the encoding process by the process of the encoding step.
The first data may include at least a part of the quantization accuracy information of the encoding process by the process of the encoding step.
The first data may include at least a part of information representing the number of quantization units of the encoding process by the process of the encoding step.
The first data may be data indicating a predetermined numerical value, and the second data is a value obtained by minimizing the numerical value of the first data.
The second data can be obtained by replacing at least part of the first data with random data.
[0060]
  The data converter of the present invention is included in the first data string., At least part of the data used to reproduce the predetermined informationThe first data, When used for reproducing the predetermined information, the reproduction quality of the predetermined information is deteriorated as compared with the case of the first data string.The first replacement means for replacing with the second data is included in a different area from the first data., At least part of the data used to reproduce the predetermined informationThird dataAt least some ofThe, Control information referred to when reproducing the second data stringReplace with 4th dataAt the same time, the other part is a non-reference area that is not referred to as reproduction data when the second data string is reproduced.A second replacing unit; a first replacing unit; and a generating unit configured to generate a second data string using the data generated by the second replacing unit.PrepareIt is characterized by that.
[0062]
  The first program of the present invention is included in the first data string.It is at least a part of data used for reproducing predetermined informationThe first data, In order to reproduce the reproduction quality of the predetermined informationThe first replacement step for replacement with the second data and the first data are included in different areas., At least part of the data used to reproduce the predetermined informationThird dataAt least some ofThe, Control information referred to when reproducing the second data stringReplace with 4th dataAt the same time, the other part is a non-reference area that is not referred to as reproduction data when the second data string is reproduced.A second replacement step, a process of the first replacement step, and a generation step of generating a second data string using the data generated by the process of the second replacement step.IncludeIt is characterized by that.
[0063]
  The data reproduction method of the present inventionThe first data string includes first data that is at least a part of data used to reproduce predetermined information, and is included in a different area from the first data. Third data that is at least a part of data used for reproducing the predetermined information is further included, and the second data string includes the first data in the first data string. Of the third data in the first data string is replaced with the second data that deteriorates the reproduction quality of the predetermined information as compared with the case of the first data string. Is replaced with the fourth data, which is control information referred to when the second data string is reproduced, and the other part is reproduced data when the second data string is reproduced. As Has been set to non-reference region which is not irradiation,Contained in the second data column4th dataA reproduction control step for controlling the reproduction of the second data string based onIncludeIt is characterized by that.
[0064]
  4th dataIs information for designating whether or not the corresponding frame can be reproduced when the reproduction of the second data string is controlled by the process of the reproduction control step.includingCan be.
[0067]
  4th dataCan include at least a part of the first data replaced with the second data.
[0068]
  4th dataWhen the reproduction of the second data string is controlled by the process of the reproduction control step, the second data is restored using at least a part of the first data included in the control information. It may be assumed that information indicating whether or not is possible is further included.
[0074]
  Of the second data columnAn inverse transformation step for inverse transformation of the frequency component, and a decoding step for decoding the second data string inversely transformed by the processing of the inverse transformation step may be further provided.it can.
The third data may include at least a part of the spectral coefficient information of the frequency component converted by the processing of the frequency component conversion step.
The first data may include at least a part of the normalization coefficient information of the encoding process by the process of the encoding step.
The first data may include at least a part of the quantization accuracy information of the encoding process by the process of the encoding step.
The first data may include at least a part of information indicating the number of quantization units of the encoding process by the process of the encoding step.
[0075]
  The first data may be data indicating a predetermined numerical value, and the second data may be data obtained by minimizing the numerical value of the first data.
  The second data can be obtained by replacing at least part of the first data with random data.
[0080]
  The second program of the present invention is:The first data string includes first data that is at least a part of data used to reproduce predetermined information, and is included in a different area from the first data. Third data that is at least a part of data used for reproducing the predetermined information is further included, and the second data string includes the first data in the first data string. Of the third data in the first data string is replaced with the second data that deteriorates the reproduction quality of the predetermined information as compared with the case of the first data string. Is replaced with the fourth data, which is control information referred to when the second data string is reproduced, and the other part is reproduced data when the second data string is reproduced. As When those which are the non-reference region which is not irradiation,Contained in the second data column4th dataA reproduction control step for controlling the reproduction of the second data string based onIncludeIt is characterized by that.
[0081]
  The data restoration method of the present inventionThe first data string includes first data that is at least a part of data used for reproducing predetermined information, and the second data string is the first data string. When the first data is used to reproduce the predetermined information, the first data is replaced with second data that deteriorates the reproduction quality of the predetermined information as compared with the case of the first data string. 2Data columnFirstBased on the acquisition control step for controlling the acquisition of the third data string including information necessary for restoring the data string, and the third data string whose acquisition is controlled by the processing of the acquisition control step, the second Data columnTo the first data columnAnd the third data string includes first data for restoring the second data string, and in the process of the restoration step, the acquisition is controlled by the process of the acquisition control step The first data included in the third data column isSecondBy replacing the second data contained in the data column,FirstThe data string is restored, and part of the second data isSecondThe control information is referred to when the data string is reproduced.
[0101]
  In the data conversion method and data conversion apparatus of the present invention, and the first program, they are included in the first data string., At least part of the data used to reproduce the predetermined informationThe first data, When used for reproducing the predetermined information, the reproduction quality of the predetermined information is deteriorated as compared with the case of the first data string.Replaced with the second data and included in a different area from the first data, At least part of the data used to reproduce the predetermined informationThird dataAt least some ofBut, Control information referred to when reproducing the second data stringReplaced by the fourth dataAt the same time, the other part is a non-reference area that is not referred to as reproduction data when the second data string is reproduced.Using the data generated by data replacement, the second data string is generatedBe done.
[0102]
  In the data reproduction method and the second program of the present invention,The first data string includes first data that is at least a part of data used to reproduce predetermined information, and is included in a different area from the first data. Third data that is at least a part of data used for reproducing the predetermined information is further included, and the second data string includes the first data in the first data string. Of the third data in the first data string is replaced with the second data that deteriorates the reproduction quality of the predetermined information as compared with the case of the first data string. Is replaced with the fourth data, which is control information referred to when the second data string is reproduced, and the other part is reproduced data when the second data string is reproduced. As Has been set to non-reference region which is not irradiation,Contained in the second data column4th dataThe second data string is played back based onBe done.
[0103]
  In the data restoration method of the present invention,The first data string includes first data that is at least a part of data used for reproducing predetermined information, and the second data string is the first data string. When the first data is used to reproduce the predetermined information, the first data is replaced with second data that deteriorates the reproduction quality of the predetermined information as compared with the case of the first data string. 2Data columnFirstA third data string including information necessary for restoring the data string is acquired, and the second data string is obtained based on the third data string.To the first data columnIs restored and the third data column isFirstIncluding the first data for restoring the data string, the first data included in the third data string,SecondBy replacing the second data contained in the data column,FirstThe data string is restored and part of the second data isSecondThis is control information referred to when reproducing a data string.
[0105]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[0106]
Here, high efficiency coding is performed by receiving input of a digital signal such as an audio PCM signal and performing band division coding (SBC), adaptive transform coding (ATC) and adaptive bit allocation. The case will be described. Adaptive transform coding is a coding method in which bit allocation is adapted based on discrete cosine transform (DCT) or the like, and an input signal is converted into a spectrum signal for each time block, and for each predetermined band, Normalizes spectrum signals together, that is, normalizes coefficients that approximate the maximum signal component, divides each signal component, and then quantizes and encodes the signal with the quantization accuracy determined by the nature of the signal. is there.
[0107]
FIG. 1 is a block diagram illustrating a configuration example of an encoding device 1 that receives an input of an acoustic waveform signal and creates trial listening data.
[0108]
The converter 11 receives an acoustic waveform signal, converts it into a signal frequency component, and outputs it to the signal component encoder 12. The signal component encoding unit 12 encodes the input signal frequency component and outputs the encoded signal frequency component to the code string generation unit 13. The code sequence generation unit 13 generates a code sequence from the signal frequency component encoded by the signal component encoding unit 12 and outputs the code sequence to the trial listening data generation unit 14. The trial-listening data generation unit 14 performs predetermined processing such as rewriting of normalization coefficient information and insertion of control information on the code string input from the code string generation unit 13 and can be reproduced with high sound quality. (Original data) is converted into trial data, and additional data (restoration data) corresponding to the trial data that is sold to a user who wants to reproduce the original data is generated and output.
[0109]
FIG. 2 is a block diagram showing a more detailed configuration of the conversion unit 11.
[0110]
The acoustic waveform signal input to the conversion unit 11 is divided into two bands by the band division filter 21, and the respective signals are output to the forward spectrum conversion units 22-1 and 22-2. The forward spectrum conversion units 22-1 and 22-2 convert the input signal into a spectrum signal component using, for example, MDCT, and outputs the signal to the signal component encoding unit 12. The signals input to the forward spectrum conversion units 22-1 and 22-2 are ½ of the bandwidth of the signal input to the band division filter 21, and the signal input is also decimated to ½. ing.
[0111]
In the conversion unit 11 of FIG. 2, the signal divided into two bands by the band division filter 21 has been described as being converted into a spectrum signal component using MDCT, but the input signal is converted into a spectrum signal component. Any conversion method may be used. For example, the input signal may be converted into a spectrum signal component using MDCT without dividing the band. Alternatively, the forward spectrum conversion units 22-1 and 22-2 may convert an input signal into a spectrum signal using DCT or DFT.
[0112]
  By using a so-called band division filter, it is possible to divide an input signal into band components, but MDCT, DCT, orDFTIt is preferable to perform spectral conversion using
[0113]
In FIG. 2, the input acoustic waveform signal has been described as being divided into two bands by the band division filter 21, but it goes without saying that the number of band divisions may not be two. . Information indicating the number of band divisions in the band division filter 21 is output to the code string generation unit 13 via the signal component encoding unit 12.
[0114]
FIG. 3 is a diagram showing the absolute value of the spectrum signal obtained by the MDCT obtained by the conversion unit 11 converted into a power level. The acoustic waveform signal input to the conversion unit 11 is converted into, for example, 64 spectrum signals for each predetermined time block. These spectral signals are processed by the signal component encoding unit 12 by the processing described later, for example, 16 pieces [1] to [16] as shown by 16 frames surrounded by solid lines in the figure. Quantization and normalization are performed for each band. A set of spectrum signals divided into 16 bands, that is, a set of spectrum signals to be quantized and normalized is a quantization unit.
[0115]
By changing the quantization accuracy for each quantization unit based on the distribution method of the frequency components, it is possible to perform efficient coding that can minimize deterioration of sound quality heard by humans.
[0116]
FIG. 4 is a block diagram showing a more detailed configuration of the signal component encoding unit 12. Here, the frequency component encoding unit 12 separates a signal component in which energy is concentrated around a specific frequency from an input spectrum signal, in particular, a tone portion that is particularly important for hearing, A case will be described where encoding is performed separately from the components.
[0117]
The spectrum signal input from the conversion unit 11 is separated into a tone component and a non-tone component by the tone component separation unit 31, the tone component is output to the tone component encoding unit 32, and the non-tone component is converted into a non-tone component. This is output to the component encoding unit 33.
[0118]
The tone component and the non-tone component will be described with reference to FIG. For example, when the spectrum signal input to the tone component separation unit 31 is a signal as shown in FIG. 5, particularly high power level portions are separated from the non-tone components as tone components 41 to 43. The position data P1 to P3 indicating the positions of the separated tone components 41 to 43 and the width of the frequency extracted as the tone components are detected and output to the tone component encoding unit 32 together with the tone components. The
[0119]
As a method for separating tone components, for example, the method described in Japanese Patent Application No. 5-152865, WO94 / 28633, or the like filed previously by the present inventor may be used. The tone component and the non-tone component separated by this method are quantized with different numbers of bits, respectively, by the processing of the tone component encoding unit 32 and the non-tone component encoding unit 33 described later.
[0120]
The tone component encoding unit 32 and the non-tone component encoding unit 33 encode the input signals, respectively, but the tone component encoding unit 32 increases the number of quantization bits with respect to the tone component, that is, Then, the quantization accuracy is increased and quantization is performed, and the non-tone component encoding unit 33 performs quantization with respect to the non-tone components by reducing the number of quantization bits, that is, by reducing the quantization accuracy.
[0121]
For each tone component, it is necessary to add information such as the position information of the tone component and the width of the frequency extracted as the tone component, but the spectrum signal of the non-tone component should be quantized with a small number of bits. Is possible. In particular, when the acoustic waveform signal input to the encoding device 1 is a signal in which energy is concentrated in a specific spectrum, this method makes it possible to hardly perceive auditory degradation. In addition, it is possible to effectively encode at a high compression rate.
[0122]
FIG. 6 is a block diagram showing a more detailed configuration of the tone component encoding unit 32 of FIG.
[0123]
The normalizing unit 51 receives the spectrum signal of the tone component for each quantization unit, performs normalization, and outputs the normalized signal to the quantization unit 52. The quantization accuracy determination unit 53 calculates the quantization accuracy with reference to the input quantization unit, and outputs the calculation result to the quantization unit 52. Since the input quantization unit is a tone component, the quantization accuracy determination unit 53 calculates the quantization accuracy so as to increase the quantization accuracy. The quantization unit 52 quantizes the normalization result input from the normalization unit 51 with the quantization accuracy determined by the quantization accuracy determination unit 53 to generate a code, and in addition to the generated code , Encoding information such as normalization coefficient information and quantization accuracy information is output.
[0124]
The tone component encoding unit 32 also encodes and outputs the position information of the tone component input together with the tone component together with the tone component.
[0125]
FIG. 7 is a block diagram showing a more detailed configuration of the non-tone component encoding unit 33 of FIG.
[0126]
The normalization unit 54 receives the non-tone component spectrum signal for each quantization unit, performs normalization, and outputs the normalized signal to the quantization unit 55. The quantization accuracy determination unit 56 calculates the quantization accuracy with reference to the input quantization unit and outputs the calculation result to the quantization unit 55. Since the input quantization unit is a non-tone component, the quantization accuracy determination unit 56 calculates the quantization accuracy so that the quantization accuracy is lowered. The quantization unit 55 quantizes the normalization result input from the normalization unit 54 with the quantization accuracy determined by the quantization accuracy determination unit 56 to generate a code, and in addition to the generated code , Encoding information such as normalization coefficient information and quantization accuracy information is output.
[0127]
Compared to the above-described encoding method, it is possible to further increase the encoding efficiency. For example, variable length coding is performed. A relatively short code length is assigned to a frequency signal that is quantized, and a relatively long code length is assigned to an infrequent one. By assigning, the encoding efficiency can be further increased.
[0128]
Then, the code string generation unit 13 in FIG. 1 records, for example, on the recording medium from the code of the signal frequency component output by the signal component encoding unit 12 or other information processing apparatus via the data transmission path. Are generated, that is, a code string composed of a plurality of frames is generated and output to the audition data generation unit 14. The code string generated by the code string generator 13 is audio data that can be reproduced with high sound quality by a normal decoder. FIG. 8 shows a frame format of audio data that can be reproduced with high sound quality generated by the code string generator 13.
[0129]
A fixed-length header including a synchronization signal is arranged at the head of each frame. In the header, the number of band divisions of the band division filter 21 of the conversion unit 11 described with reference to FIG.
[0130]
In each frame, tone component information regarding the separated tone components is recorded following the header. In the tone component information, the number of tone components, the tone width, and quantization accuracy information of quantization applied to the tone components by the tone component encoding unit 32 described with reference to FIG. 6 are recorded. Subsequently, as the data of the tone components 41 to 43, the respective normalization coefficients, tone positions, and spectrum coefficients are recorded. Here, for example, it is assumed that the normalization coefficient of the tone component 41 is 30, the normalization coefficient of the tone component 42 is 27, and the normalization coefficient of the tone component 43 is 24.
[0131]
Then, the non-tone component information is described following the tone component information. The non-tone component information includes the number of quantization units (16 in this case), quantization accuracy information of quantization applied to the non-tone components by the tone component encoding unit 33 described with reference to FIG. Normalization coefficient information and spectral coefficient information of each quantization unit are recorded. In the normalization coefficient information, a value from 46 in the lowest quantization unit [1] to a value 8 in the highest quantization unit [16] is recorded for each quantization unit. Here, it is assumed that a value proportional to the dB value of the power level of the spectrum signal is used as the normalization coefficient information. Further, when the length of the content frame is a fixed length, an empty area may be provided after the spectrum coefficient information.
[0132]
FIG. 9 is a block diagram showing a more detailed configuration of the trial listening data generation unit 14 of FIG.
[0133]
The audition data generation processing control unit 61 is based on setting data such as the audition area of the audition data and the audition band of the audition area, which are input from an external operation input unit (not shown). The restriction processing unit 63 is controlled. The trial listening area is a portion of the trial listening data that can be reproduced with low quality without using additional data. For example, a rust portion is set as the trial listening area. A plurality of audition areas may be set in the audition data. The trial listening band in the trial listening area is a frequency band in which low quality reproduction is performed. For example, by designating only a part of the quantization units in the spectrum data described with reference to FIG. 5, only a certain range of frequency bands can be reproduced, and the quality of reproduced audio is lowered. ing.
[0134]
The audition area determination unit 62 determines whether or not the input frame is within the audition area for each frame described with reference to FIG. 8 according to the control of the audition data generation processing control unit 61. The result is output to the band limitation processing unit 63 and the control information insertion unit 64. As shown in FIG. 10, for example, as shown in FIG. 10, the trial area determination unit 62 converts the input frame sequence into a protected area, that is, an area where the audition is not permitted, and a trial area. Distinguish.
[0135]
Based on the signal input from the audition area determination unit 62, the band limitation processing unit 63 limits the audition band based on whether or not the input frame is a frame included in the audition area. Generate data. Information about the trial band in the trial area is input from the trial data generation processing control unit 61.
[0136]
For example, when the quantization unit [1] to the quantization unit [12] are designated as the trial listening band of the trial listening possible part, the band limitation processing unit 63, when the input frame is a frame in the trial listening area, As shown in FIG. 11, the value of the normalization coefficient information of the quantization unit [13] to the quantization unit [16] higher than the trial band is, for example, minimized to be the normalization coefficient 0. Accordingly, although effective values are described in the spectral coefficient information corresponding to the quantization units [13] to [16], the normalized coefficient information is 0 at the time of reproduction. Strictly speaking, the spectrum of the portion does not become zero, but from the viewpoint of audibility, it is substantially equal to zero. Accordingly, the spectral coefficient information on the higher frequency side than the position indicated by Ad in the figure is synonymous with not being referred to.
[0137]
Similar to the non-tone component, the normalization coefficient value of the tone component that is outside the listening band is also minimized, for example, to 0, thereby minimizing the spectrum signal of the corresponding tone component during playback. (Substantially change to a value equivalent to 0).
[0138]
FIG. 12 shows a spectrum signal when the sample data shown in FIG. 11 is reproduced. Since the normalization coefficient information of the quantization unit [13] to the quantization unit [16] is changed to 0, the corresponding spectrum signal is minimized. Similarly, the corresponding spectrum signal is minimized for the two tone components included in the quantization unit [13] to the quantization unit [16]. That is, when the trial listening data is decoded and reproduced, only the narrow-band spectrum signal is reproduced in the portion where the trial listening is possible.
[0139]
In this way, when the trial listening area of the trial listening data is reproduced, only narrow-band data is reproduced, so that data with lower quality than the original data described with reference to FIG. 8 is reproduced.
[0140]
Furthermore, the band limitation processing unit 63 sets the values of all the normalization coefficient information of the quantization units [1] to [16] of the frame corresponding to the part where the trial listening is impossible to 0. Thereby, the data of the frame (protection area) that cannot be listened to is silent even if it is reproduced.
[0141]
The band limitation processing unit 63 uses the replacement data so that the decoded data length when the sample data is decoded does not become longer than the decoded data length when the original data is decoded. Generate.
[0142]
The control information inserting unit 64 prevents only unsound reproduction such as a case where the trial listening part suddenly flows after the silent state continues when the trial listening data is reproduced. Control information indicating whether or not playback is possible is inserted into all frames so that the frames are continuously played back. The control information insertion unit 64 controls the control information into a region of spectrum coefficient information that is not referred to, that is, a region that is ignored when the sample data is reproduced because the reproduction quality is limited in accordance with the information input from the sample region determination unit 62. Is output to the trial data generator 66. Then, the control information insertion unit 64 adds the true spectrum coefficient information corresponding to the portion where the control information is inserted, and information indicating the position where the control information is inserted, if necessary, to the additional frame generation unit 65. Output to.
[0143]
FIG. 13 shows a frame format when control information is inserted.
[0144]
As described with reference to FIG. 11, when the value of the normalization coefficient information of the quantization unit [13] to the quantization unit [16] higher than the trial band is changed to 0, the quantization unit [ 13] to the quantization unit [16], the spectral coefficient information on the higher frequency side than the position indicated by Ad in the figure is synonymous with that it is not referred to, so any control information is described within this range. Is possible.
[0145]
Furthermore, the control information insertion unit 64 may describe random dummy data or the like at a position where control information is not described within the range of spectral coefficient information on the higher frequency side than the position indicated by Ad in the figure. . In this case, the control information insertion unit 64 sends the true spectrum coefficient information corresponding to the portion where the dummy data is described, and information indicating the position where the dummy data is inserted, if necessary, to the additional frame generation unit 65. Output to.
[0146]
Here, it has been described that the control information is inserted at a position corresponding to the spectral coefficient information that is not referred to when the sample data is reproduced, but if the part is not referred to when the sample data is reproduced, the position at which the control information is inserted is It may be other than the position corresponding to the spectral coefficient information.
[0147]
An example of the format of the control information is shown in FIG.
[0148]
In the control information, a playback device that receives input of audition data and reproduces it determines whether the corresponding frame is a frame for which audition is permitted or a frame for which audition is not permitted. As information, a trial listening permission flag, a trial listening start flag, and a trial listening end flag are described.
[0149]
The trial listening permission flag is set by a person who has the copyright of the content, and is for distinguishing between a frame that prohibits the trial listening and a frame that is permitted. In the playback device that plays the audition data, the audition permission flag is referred to, and if the audition is permitted, the normal playback process is performed, and if the audition is not permitted, the next audition frame is not processed. The process for is started. As a result, the playback process is skipped for frames for which trial listening is prohibited, so that silent playback can be prevented even when there are frames that cannot be auditioned before and after the trial listening area.
[0150]
The audition start flag and the audition end flag are information set in a predetermined time after the start position of the audition area and a predetermined time before the end position for a group of the audition area. Is set to 3 seconds, the audition start flag is set so that it can be discriminated in the trial frame that is played for 3 seconds from the start of the trial area, and the trial frame is played within 3 seconds before the end of the trial area. The audition end flag is set so that it can be discriminated.
[0151]
Thereby, for example, when the trial listening data is played back by the playback device, playback processing such as fade-in and fade-out can be performed before and after the trial listening area. In other words, when there are a plurality of audition areas, in the original original data, a plurality of audition areas that should be reproduced at different locations are continuously reproduced at the same volume, whereas a plurality of audition areas are reproduced. At each start and end, playback processing such as fade-in and fade-out can be performed before and after the trial listening area, so that the trial listening data can be more uncomfortable for the user and can increase purchase motivation. it can.
[0152]
In particular, when the spectral coefficient information is variable-length encoded and the variable-length code is sequentially described from the low-frequency side to the high-frequency side in the spectral coefficient information description area, Since the control information is described in the area, a part of the middle-range variable length code is lost, and therefore the high-frequency data including that part cannot be decoded at all. That is, since it becomes difficult to restore the spectral coefficient information related to the original data included in the trial listening data without using additional data, the safety of the trial listening data is enhanced.
[0153]
As described above, when a part of data such as a normalization coefficient is replaced with 0 or control information, the estimation of true data for the replaced data is performed by using an encryption key having a relatively short key length. Compared to deciphering, it is very difficult. In addition, if the trial listening data is illegally altered, sound quality is deteriorated. Therefore, it is very difficult for a user who is not permitted to audition the original data to guess the original data based on the audition data, and the rights of the authors and distributors of the content data can be further protected. It becomes possible.
[0154]
Also, in the unlikely event that the true data for the replaced data is inferred in some sample data, unlike the case where the encryption algorithm is decrypted, the damage will not spread to other content. The content data encrypted by using a specific algorithm is safer than distributing it as trial listening data.
[0155]
In the above, the true normalization coefficient information and the true spectrum coefficient information replaced by the band limitation processing unit 63 and the control information insertion unit 64 are supplied to the additional frame generation unit 65 described later and described in the additional frame. Since the band limitation processing unit 63 may change values such as quantization accuracy information and the number of quantization units to 0 or random dummy data in addition to the normalization coefficient information, for example, which will be described later. The additional frame generation unit 65 receives, for example, information indicating values such as quantization accuracy information and the number of quantization units changed by the band limitation processing unit 63, and describes the information in the additional frame.
[0156]
However, when the normalization coefficient information is changed and when the quantization accuracy information is changed, it is difficult to illegally infer the original data from the trial data without using additional data, that is, the safety of the trial data. Strength will be different. For example, when a bit allocation algorithm that calculates quantization accuracy information based on normalization coefficient information is employed when generating original data, even if only the quantization accuracy information is changed, the normalization coefficient There is a risk that true quantization accuracy information can be estimated using information.
[0157]
On the other hand, even if only the normalization coefficient information is changed, it is difficult to estimate the normalization coefficient information from the quantization accuracy information, and thus it can be said that the safety strength of the trial listening data is high. It should be noted that by changing both values of the normalization coefficient information and the quantization accuracy information, the risk of illegally guessing the original data is further reduced. Further, the value of the normalization coefficient information or the quantization accuracy information may be selectively changed according to the content frame of the trial listening data.
[0158]
The additional frame generation unit 65 is an original to be purchased when the user who listened to the audition data listens to the music with high sound quality from the information on the normalized normalization coefficient or the like whose band is limited by the band limitation processing unit 63 An additional frame constituting additional data for data restoration is generated.
[0159]
FIG. 15 shows a format of the generated additional frame. As described with reference to FIG. 13, when the quantization unit [1] to the quantization unit [12] are selected as the trial listening band, the quantization unit [13] to the quantization in each frame of the trial listening data. The normalization coefficients of the tone component and the non-tone component corresponding to the four quantization units of the unit [16] are replaced with 0. In addition, a part of the spectrum coefficient information that is not referred to is replaced with control information. The additional frame generation unit 65 receives supply of information indicating the original values (true normalization coefficient information and true spectrum coefficient information) of the changed normalization coefficient information and spectrum coefficient information from the band limitation processing unit 63. The additional frame in FIG. 15 is generated.
[0160]
In the additional frame, additional information corresponding to the tone component and additional information corresponding to the non-tone component are described. FIG. 15 shows a case where quantization unit [1] to quantization unit [12] are selected as the trial band. Additional information corresponding to tone components includes the number of tone components minimized (here, two components are minimized), the position of the normalization coefficient information of each tone component on the audition frame, and the true normality. The conversion factor information is described. Further, as additional information corresponding to the non-tone component, the number of normalized coefficients of the non-tone component minimized (here, four components are minimized) and information indicating the head position of the normalized coefficient ( For example, the number 13 of the first quantization unit whose normalization coefficient is changed to 0 may be used), the rewritten true normalization coefficient, and the true spectral coefficient information of the portion rewritten in the control information Has been.
[0161]
Here, the position information of the true spectral coefficient information of the portion rewritten with the control information is not described in the generated additional frame, but the portion of the trial listening frame where the spectral coefficient information is rewritten with the control information is not referred to. If the head of the part to be the spectral coefficient information is obtained, the position of the true spectral coefficient information of the part rewritten by the control information is obtained based on the information of the region of the quantization unit in which the normalization coefficient is replaced with 0. Is possible. When the position to be rewritten to the control information is set to a position different from the head of the part that becomes the non-referenced spectral coefficient information, it is necessary to describe the position information of the true spectral coefficient information of the part rewritten to the control information in the additional frame. is there.
[0162]
In addition, when the band limitation processing unit 63 changes values such as quantization accuracy information and the number of quantization units other than the normalization coefficient information and the spectrum coefficient information, the additional frame generation unit 65 In response to the input of information indicating values such as quantization accuracy information and the number of quantization units changed by 63, information on their true values is written in the additional frame.
[0163]
Further, part or all of the normalization coefficient information described in the additional frame may be described in the control information. An example of the format of the control information in that case is shown in FIG.
[0164]
The control information includes all or part of the reproduction control information to be referred to when reproducing the sample data and the normalization coefficient information replaced to limit the band. In FIG. 16, all the information regarding the non-tone component is described in the normalization coefficient information described in the additional frame, but the information described here is, for example, a part of the non-tone component. Even normalization coefficient information, all replaced normalization coefficient information including tone components may be used, or when information other than normalization coefficient information is replaced when generating audition data Such information may be described.
[0165]
Further, in the reproduction control information, a reproducible flag is described in addition to the reproduction permission flag, the trial listening start flag, and the trial listening end flag described with reference to FIG.
[0166]
The reproducible flag indicates whether or not reproduction using normalization coefficient information described in the control information is possible. The playback device that has received the input of the audition data refers to the reproducible flag of the control information in the audition frame, and when the normalization coefficient information is available, the normalization coefficient information in the control information is converted into the original information in the audition frame. It is possible to restore the position and perform the reproduction process. Also, the playback device that has received the input of the audition data plays back without using the normalization coefficient information in the control information when the playable flag indicates that the normalization coefficient information cannot be used. Process.
[0167]
In the reproducible flag, it is also possible to specify the version and type of the playback device, for example, in a playback device that satisfies the specified version, using the normalization coefficient information included in the control information, A part of the audition data can be restored and the playback process can be performed. On the other hand, in the playback device of a version other than the specified, the normalization coefficient information included in the control information must be used in the playback process. I can't.
[0168]
Also, by describing a part of the information to be added to the additional data in the control information, the data capacity of the additional data can be reduced, so that the user records the additional data on the recording medium using, for example, MMK. When trying to do so, the processing time can be shortened, or the communication time can be shortened when additional data is to be obtained through download processing.
[0169]
When the control information of FIG. 16 is inserted into the trial frame, the band limitation processing unit 63 uses the control information instead of the additional frame generation unit 65 as a whole or a part of the normalized coefficient information replaced to limit the band. The data is output to the insertion unit 64. The control information insertion unit 64 generates the control information of FIG. 16 using the true normalization coefficient information input from the band limitation processing unit 63, and inserts it into the trial listening frame.
[0170]
In addition, as described with reference to FIG. 16, a part of the information described in the additional frame, for example, when the true normalization coefficient information of the non-tone part is described in the control information, the corresponding part Is not described in the additional frame.
[0171]
The sample data generation unit 66 generates a sample data header, adds the generated header to the encoded frame sequence of the input sample data, and generates and outputs sample data. The header of the trial listening data includes, for example, information such as a content ID for identifying the content, a content reproduction time, a content title, or encoding method information.
[0172]
The additional data generation unit 67 generates a header of the additional data, adds the generated header of the additional data to the encoded frame sequence of the input additional data, and generates and outputs the additional data. In the header of the additional data, a content ID for identifying the content and making it correspond to the trial listening data, the playback time of the content, and information on the encoding method as necessary are described.
[0173]
In this way, since the control information is described so that the trial listening area of the trial listening data is continuously played back, it is possible to prevent the protected area from being played back silently, and to listen to the trial listening data. It is possible to prevent the user who has done this from feeling uncomfortable.
[0174]
In addition, by describing various information related to the playback of sample data in the control information, for example, processing such as fade-in and fade-out can be performed at the beginning and end of the sample area, so that it is different from the original part. Even when multiple preview areas that should be played are played back continuously, the user who auditioned the audition data can understand the continuous and non-continuous parts of the original audition area, The reproduction quality of the trial listening data can be changed depending on the version of the reproduction apparatus.
[0175]
Furthermore, by including the true data that was replaced when the sample frame was generated in the control information, the data volume of the additional data can be reduced, or the playback quality of the sample data depending on the type and version of the playback device. Can be changed.
[0176]
Using the trial listening data and the additional data generated by the trial listening data generation unit 14, the original data can be restored by the processing described later.
[0177]
Next, the trial listening data generation process will be described with reference to the flowcharts of FIGS. 17 and 18.
[0178]
In step S1, the audition data generation processing control unit 61 acquires a setting value of the audition band of the audition area of the audition data input from an operation input unit (not shown) or the like. Here, as described with reference to FIG. 11 and FIG. 12, the description will be made assuming that quantization unit [1] to quantization unit [12] are set as trial listening bands. The audition data generation processing control unit 61 supplies the setting value of the audition band to the band limitation processing unit 63.
[0179]
In step S2, the audition data generation processing control unit 61 acquires information specifying an audition area of the audition data input from an operation input unit (not shown) or the like. Here, description will be made assuming that the trial listening area 1 composed of the fourth and fifth frames and the trial listening area 2 composed of the ninth and tenth frames described with reference to FIG. 10 are designated as the trial listening areas. The audition data generation processing control unit 61 supplies information specifying the audition area to the audition area determination unit 62.
[0180]
In step S3, the trial listening area determination unit 62 and the band limitation processing unit 63 input any of the frames included in the frame sequence corresponding to the original data, that is, the frame capable of high-quality reproduction described with reference to FIG. receive.
[0181]
In step S <b> 4, the trial area determination unit 62 determines whether the input frame is a trial frame (a frame included in the trial area) based on the information supplied from the trial data generation processing control unit 61. .
[0182]
If it is determined in step S4 that the input frame is a trial listening frame, the trial listening area determination unit 62 supplies the determination result to the band limitation processing unit 63 and the control information insertion unit 64 in step S5. The restriction processing unit 63 includes, for example, a portion other than the band specified by the set value of the trial band supplied from the trial data generation processing control unit 61 in the normalization coefficient information of the tone component of the input frame, for example, Change to a value such as 0.
[0183]
In step S <b> 6, the band limitation processing unit 63 includes, for example, a portion other than the band specified by the set value of the trial band supplied from the trial data generation processing control unit 61 in the normalization coefficient information of the non-tone component. 11, the trial frame described with reference to FIG. 11 is generated and output to the control information insertion unit 64, and the true normalization coefficient information replaced in steps S 5 and S 6 and its position Are output to the additional frame generation unit 65, or a part of the information is output to the control information insertion unit 64 and the rest is output to the additional frame generation unit 65.
[0184]
In step S7, the control information insertion unit 64 describes the control information indicating whether or not the trial listening is possible in part of the non-referenced spectral coefficient information corresponding to the normalized coefficient information of the non-tone component rewritten in the process of step S6. The sample listening frame described with reference to FIG. 13 is generated. Here, even if the control information described is only for the reproduction control information as described with reference to FIG. 14, a part of the information described in the additional data as described with reference to FIG. It does not matter if it contains any. The control information insertion unit 64 outputs the generated trial listening frame to the trial listening data generation unit 66, and outputs the true spectral coefficient information replaced with the control information in step S7 to the additional frame generation unit 65.
[0185]
Here, when the spectrum coefficient information is variable-length encoded, control information is used such that the bit length when the control information is decoded is shorter than the bit length when the true spectrum coefficient information is decoded. As a result, it is possible to prevent the frame length of the code string from being exceeded in the decoding process described later.
[0186]
Further, the control information insertion unit 64 may replace dummy data in addition to the control information with part of the spectrum coefficient information that is not referred to. The dummy data to be replaced with the spectrum coefficient information may be all set to the value 0, or the value 1 and the value 0 may be appropriately mixed.
[0187]
In step S8, the additional frame generation unit 65 is added when the user who auditioned the audition data listens to the music with high sound quality based on the signals input from the band limitation processing unit 63 and the control information insertion unit 64. Data for an additional frame constituting the data is generated.
[0188]
In step S4, when it is determined that the input frame is not a trial listening frame, that is, a protection frame (a frame included in the protection area), in step S9, the trial listening area determination unit 62 limits the determination result to a band limit. The data is supplied to the processing unit 63 and the control information insertion unit 64. The band limitation processing unit 63 changes all the tone component normalization coefficient information to the minimum value (value 0).
[0189]
In step S10, the band limitation processing unit 63 changes all the normalization coefficient information of the non-tone components to the minimum value (value 0) and outputs it to the control information insertion unit 64, and replaces it in step S9 and step S10. All of the obtained normal normalization coefficient information and various kinds of information indicating the position thereof are output to the additional frame generation unit 65 or a part thereof is output to the control information insertion unit 64 and the other is generated as additional frames To the unit 65.
[0190]
In step S11, the control information insertion unit 64 describes the control information indicating that the trial listening is impossible in a part of the non-referenced spectral coefficient information corresponding to the normalized coefficient information of the non-tone component rewritten in the process of step S10. . The control information insertion unit 64 outputs the generated trial listening frame to the trial listening data generation unit 66, and outputs the true spectral coefficient information replaced with the control information in step S11 to the additional frame generation unit 65.
[0191]
Similarly, when the spectral coefficient information is variable-length encoded, the bit length when the control information is decoded is shorter than the bit length when the true spectral coefficient information is decoded. Replacement is performed using control information. Furthermore, replacement may be performed using dummy data in addition to the control information.
[0192]
In step S12, the additional frame generation unit 65 is added when the user who auditioned the audition data listens to the music with high sound quality based on the signals input from the band limitation processing unit 63 and the control information insertion unit 64. Data for an additional frame constituting the data is generated.
[0193]
After the process of step S8 ends, or after the process of step S12 ends, in step S13, the trial listening data generation unit 64 determines whether or not the processed frame is the final frame. If it is determined in step S13 that the processed frame is not the final frame, the process returns to step S3, and the subsequent processing is repeated.
[0194]
In step S13, when it is determined that the processed frame is the final frame, in step S14, the trial data generation unit 66 generates a header of the trial data and adds the sample data to the trial frame sequence. Generate and output.
[0195]
In step S15, the additional data generation unit 67 generates a header of additional data using the input information, adds the header to the additional frame sequence, generates and outputs the additional data, and the process ends.
[0196]
The process described with reference to the flowcharts of FIGS. 17 and 18 generates audition data in which only the audition area is reproduced with low quality, and additional data for restoring the original data from the audition data.
[0197]
In order to stimulate the user's willingness to purchase, the data in the trial listening area may have the same high sound quality as the original data without band limitation. In that case, without limiting the bandwidth of the data in the trial listening area, the frame in the trial listening area of the original data may be copied to the trial listening frame as it is, and an additional frame for that portion may not be generated.
[0198]
The signal component encoding unit 12 of the encoding apparatus 1 has been described as separating the tone component and the non-tone component and performing the encoding separately when encoding the input signal. By using the non-tone component encoding unit 33 in FIG. 7 instead of the encoding unit 12, the input signal may be encoded without separating the tone component and the non-tone component.
[0199]
FIG. 19 shows the format of an original data frame of high sound quality generated by the code string generator 13 when the input signal is not separated into tone components and non-tone components. As in the case described with reference to FIG. 8, a fixed-length header including a synchronization signal is arranged at the beginning of the original data frame. In the header, the number of band divisions of the band division filter 21 of the conversion unit 11 described with reference to FIG. Following the header, the number of quantization units (here 16), quantization accuracy information of quantization performed by the non-tone component encoding unit 33, normalization coefficient information of each of the 16 quantization units, and spectral coefficient information Is recorded. The normalization coefficient information is recorded for each quantization unit from a value of 46 in the lowest quantization unit [1] to a value of 8 in the highest quantization unit [16]. Further, when the length of the content frame is a fixed length, an empty area may be provided after the spectrum coefficient information.
[0200]
FIG. 20 shows the format of the audio data of the trial part generated by the trial data generation unit 14 that has received the input of the original data frame described with reference to FIG. For example, when the quantization unit [1] to the quantization unit [12] are designated as the trial listening band of the trial listening possible portion, the quantization on the higher frequency side than the trial listening band is performed as in the case described with reference to FIG. The value of the normalization coefficient information of units [13] to [16] is set to 0. Accordingly, although effective values are described in the spectral coefficient information corresponding to the quantization units [13] to [16], the normalized coefficient information is 0 at the time of reproduction. The portion of the spectrum is minimized. And the control information demonstrated using FIG. 14 or FIG. 16 is described in a part of spectrum coefficient which is not referred.
[0201]
FIG. 21 shows an additional frame generated by the additional frame generation unit 65 of the trial listening data generation unit 14 that has received the input of the original data frame described with reference to FIG. Here, an additional frame when only the reproduction control information described with reference to FIG. 14 is described as control information will be described. The additional frame includes the number of normalized coefficients of the minimized quantization frame (here, the four components are minimized), the position of the head of the normalized coefficient, the rewritten true normalization coefficient, and the control The true spectral coefficient information of the rewritten portion is written in the information.
[0202]
As described with reference to FIGS. 19 to 21, even when tone components are not separated, the same processing is performed, and the audition data in which only the audition area is reproduced and only the audition area is reproduced, and the original from the audition data is reproduced. Additional data for restoring the data is generated.
[0203]
The audition data generated in this way is distributed to the user via the Internet or the like, or recorded and distributed on various recording media owned by the user by the MMK provided in the store or the like. The user who likes the playback of the trial listening data can obtain additional data, for example, by paying a predetermined fee to a content data distribution company. The user can restore the original data by using the trial listening data and the additional data, decode and reproduce it, or record it on the recording medium.
[0204]
Next, a description will be given of a process in the case where the sample data is decoded and output or reproduced, or the original data is decoded and output or reproduced from the sample data and the additional frame.
[0205]
FIG. 22 is a block diagram showing the configuration of the data reproducing device 81. As shown in FIG.
[0206]
The code string decomposing unit 91 receives input of the encoded sample data, decomposes the code string, extracts the code of each signal component, and outputs the code to the code string restoring unit 93.
[0207]
The control unit 92 receives a user operation from an operation input unit (not shown), receives information indicating whether or not the input data is to be reproduced with high sound quality, receives input of additional data, and receives a code string restoration unit. 93 processing is controlled. Further, the control unit 92 supplies true coding coefficient information, true spectrum coefficient information, or the like to the code restoration unit 93 as necessary.
[0208]
Based on the control of the control unit 92, the code string restoration unit 93 outputs the input encoded frame as it is to the signal component decoding unit 94 when the input trial data is auditioned, that is, reproduced as it is. When the input audition data is restored to the original data and played back, the audition data of the audition data is based on various information such as true coding coefficient information or true spectrum coefficient information supplied from the control unit 92. The process of restoring the encoded frame to the encoded frame of the original data is executed, and the restored encoded frame of the original data is output to the signal component decoding unit 94.
[0209]
In addition, when the sample data is reproduced, the code string restoration unit 93 may restore a part or the whole of the sample data by referring to the control information. In other words, the trial data includes true coding coefficient information and the like, which is indicated by the reproducible flag, and is permitted to restore the trial data using them (for example, specification of the version of the data playback device 81, etc.) ) Is satisfied, the code string restoration unit 93 can restore a part or the whole of the trial listening data by using the true coding coefficient information included in the trial listening data.
[0210]
The signal component decoding unit 94 decodes the input sample data or the encoded frame of the original data. FIG. 23 is a block diagram showing a more detailed configuration of the signal component decoding unit 94 that decodes an encoded frame when the input encoded frame is encoded by being divided into a tone component and a non-tone component. is there.
[0211]
For example, the frame separation unit 101 receives an encoded frame as described with reference to FIG. 13 and divides it into a tone component and a non-tone component. The tone component is sent to the tone component decoding unit 102 as a non-tone component. Is output to the non-tone component decoding unit 103.
[0212]
FIG. 24 is a block diagram showing a more detailed configuration of the tone component decoding unit 102. The inverse quantization unit 111 inversely quantizes the input encoded data and outputs it to the inverse normalization unit 112. The denormalization unit 112 denormalizes the input data. That is, decoding processing is performed by the inverse quantization unit 111 and the inverse normalization unit 112, and the spectrum signal of the tone portion is output.
[0213]
FIG. 25 is a block diagram showing a more detailed configuration of the non-tone component decoding unit 103. The inverse quantization unit 121 inversely quantizes the input encoded data and outputs it to the inverse normalization unit 122. The denormalization unit 122 denormalizes the input data. That is, decoding processing is performed by the inverse quantization unit 121 and the inverse normalization unit 122, and the spectrum signal of the non-tone portion is output.
[0214]
The spectrum signal synthesis unit 104 receives the spectrum signals output from the tone component decoding unit 102 and the non-tone component decoding unit 103, synthesizes these signals, and if the original data, FIG. 5 or the audition data is used. If there is, the spectrum signal described with reference to FIG. 12 is generated and output to the inverse conversion unit 95.
[0215]
When the encoded data is not encoded by being divided into a tone component and a non-tone component, the frame separation unit 101 is omitted and the tone component decoding unit 102 or the non-tone component decoding unit 103 Decoding processing may be performed using only one of them.
[0216]
FIG. 26 is a block diagram showing a more detailed configuration of the inverse transform unit 95.
[0217]
The signal separation unit 131 separates the signal based on the number of band divisions described in the header of the input frame. Here, it is assumed that the number of band divisions is 2, and the signal separation unit 131 separates the input spectrum signal into the inverse spectrum conversion units 132-1 and 132-2.
[0218]
The inverse spectrum conversion units 132-1 and 132-2 perform inverse spectrum conversion on the input spectrum signal, and output the obtained signals of each band to the band synthesis filter 133. The band synthesis filter 133 synthesizes and outputs the input signals of the respective bands.
[0219]
A signal (for example, an audio PCM signal) output from the band synthesis filter 133 is converted into analog data by, for example, a D / A conversion unit (not shown), and reproduced and output as sound from a speaker (not shown). Further, the signal output from the band synthesis filter 133 may be output to another device via a network or the like.
[0220]
Next, the data reproduction process executed by the data reproduction device 81 in FIG. 22 will be described with reference to the flowchart in FIG.
[0221]
In step S31, the code string decomposing unit 91 receives input of the encoded frame of the trial listening data. In step S32, the code string decomposing unit 91 decomposes the input code string and outputs it to the code string restoring unit 93.
[0222]
In step S33, the code string restoration unit 93 determines, based on the signal input from the control unit 92, whether high-quality reproduction, that is, processing for restoring and reproducing the original data is executed.
[0223]
If it is determined in step S33 that high sound quality reproduction is to be executed, a code string restoration process, which will be described later, is executed in step S34 using the flowchart of FIG.
[0224]
If it is determined in step S33 that the high-quality sound reproduction is not executed, the input frame is reproduced as trial listening data. Therefore, in step S35, the code string restoration unit 93 controls the control information included in the input frame. Referring to FIG. 4, it is determined whether or not this frame is a frame that can be auditioned.
[0225]
If it is determined in step S35 that this frame is not a frame that can be auditioned, in step S36, the code string restoration unit 93 skips processing for the input frame. That is, the code string restoration unit 93 discards the input frame without outputting it to the signal component decoding unit 94, and the process proceeds to step S39.
[0226]
After the process of step S34 is completed or when it is determined in step S35 that this frame is a frame that can be auditioned, in step S37, the signal component decoding unit 94 converts the input code string into a tone component. And the non-tone component, respectively, are decoded by performing inverse quantization and inverse normalization, and the spectrum signal generated by the decoding is synthesized and output to the inverse transform unit 95.
[0227]
In step S38, the inverse conversion unit 95 performs band separation on the input spectrum signal as necessary, performs inverse spectrum conversion on each, and then performs band synthesis to inversely convert the signal into a time series signal.
[0228]
After the process of step S36 or step S38 is completed, in step S39, the control unit 92 determines whether or not the reverse frame converted by the reverse conversion unit 95 in step S38 is the final frame of the audition data. To do.
[0229]
If it is determined in step S39 that the frame is not the last frame, the process returns to step S31, and the subsequent processes are repeated. If it is determined in step S39 that the frame is the last frame, the process ends.
[0230]
The time series signal generated by the inverse conversion by the inverse conversion unit 95 may be converted to analog data by a D / A conversion unit (not shown) and reproduced and output from a speaker (not shown), or a network (not shown). It may be output to other devices via the.
[0231]
Further, when the audition data is reproduced, for example, processing such as fade-in and fade-out can be performed by referring to the audition start flag and the audition end flag of the control information as necessary.
[0232]
Note that, here, a case has been described in which the sample data that is encoded by dividing the tone component and the non-tone component, or the original data restored from the sample data, is decoded. Even when the tone component is not divided, restoration processing and reproduction processing can be performed in the same manner.
[0233]
Next, the code sequence restoration process executed in step S34 in FIG. 27 will be described with reference to the flowchart in FIG.
[0234]
In step S <b> 51, the code string restoration unit 93 acquires additional data information such as sample area information, true coding coefficient information, and true spectrum coefficient information from the control unit 92 in order to restore the code string. .
[0235]
In step S52, the code string restoration unit 93 receives an input of the frame decomposed by the code string decomposition unit 91. In step S53, the code string restoration unit 93 determines whether the input frame is a trial frame.
[0236]
When it is determined in step S53 that the input frame is a trial listening frame, in step S54, the code string restoration unit 93 performs normalization coefficient information of the tone component included in the additional frame described with reference to FIG. Is used to restore the normalization coefficient information of the tone component of the portion replaced with 0 in the trial listening frame of the trial listening data.
[0237]
In step S55, the code string restoration unit 93 uses the true value of the normalization coefficient information of the non-tone component included in the additional frame described with reference to FIG. The normalization coefficient information of the non-tone component of the part is restored.
[0238]
In step S56, the code string restoration unit 93 is replaced with the control information in the trial frame of the trial data using the true value of the spectral coefficient information of the non-tone component included in the additional frame described with reference to FIG. The spectral coefficient information of the partial non-tone components is restored, and the process returns to step S37 in FIG.
[0239]
If it is determined in step S53 that the input frame is not a trial listening frame, that is, a protected frame, in step S57, the code string restoration unit 93 determines the tone component included in the additional frame corresponding to the protected frame. Using the true value of the normalization coefficient information, the normalization coefficient information of the entire tone component replaced with 0 in the trial listening frame of the trial listening data is restored.
[0240]
In step S58, the code string restoration unit 93 uses the true value of the normalization coefficient information of the quantized frame of the non-tone component included in the additional frame corresponding to the protection frame, and replaces it with 0 in the trial data frame of the trial data. The normalization coefficient information of the entire non-tone component being restored is restored.
[0241]
In step S59, the code string restoration unit 93 uses the true value of the spectral coefficient information of the non-tone component included in the additional frame corresponding to the protection frame, and uses the true value of the spectral coefficient information of the non-tone component of the sample data. The spectral coefficient information of the non-tone component is restored, and the process returns to step S37 in FIG.
[0242]
By the process described with reference to the flowchart of FIG. 28, the original data is restored using the trial listening data and the additional data.
[0243]
In FIG. 28, the original data is restored using the additional data described with reference to FIG. 15, that is, the true normalization coefficient information is described as being all described in the additional data. In step S54 and step S55, or in step S57 and step S58, the code string restoration unit 93 adds control information in the trial frame to the control information in the case where part or all of the spectral coefficient information is described in the control information. Using the true normalization coefficient information described, the normalization coefficient information of the replaced part is restored.
[0244]
Even if the decoded trial listening data or the restored and decoded original data is reproduced using a speaker (not shown) or the like by the processing described with reference to FIGS. 22 to 28, for example, via a network or the like. You may make it output to another apparatus.
[0245]
In FIG. 27, when the audition data in which the normalization coefficient of the quantization unit 13 to the quantization unit 16 is changed to 0 is reproduced, the data in the band corresponding to the quantization unit 1 to the quantization unit 12 is reproduced. Only described as being played. On the other hand, for example, the true normalization coefficient of the quantization unit 13 and the quantization unit 14 is included in the control information of the audition frame, and the true normalization described in the control information is indicated by the playable flag. The reproduction processing may be permitted using the coefficient information, and the true normalization coefficients of the quantization unit 15 and the quantization unit 16 may be described in the additional data. When the trial listening data is reproduced, the code string restoration unit 93 restores the true normalization coefficient information of the quantization unit 13 and the quantization unit 14 based on the control information, and the quantization units 1 to Quantum It may be possible to perform a slightly higher quality reproduction process than when only the data of the band corresponding to the data conversion unit 12 is reproduced.
[0246]
When the trial listening data is normally reproduced, the code string restoration unit 93 does not refer to the normalization coefficient information included in the control data of the trial listening frame, so that the reproduction is performed by the quantization unit 1 to the quantization unit 12. Only the corresponding band data. For example, when the data reproducing apparatus is a predetermined version, or when a predetermined password is input from an operation input unit (not shown), the control unit 92 controls the code string restoration unit 93, The true normalization coefficient of the quantization unit 13 and the quantization unit 14 included in the control data of the trial listening frame is extracted, and the normalization coefficient of the corresponding part is restored. As a result, it is possible to perform a slightly higher quality reproduction process than when only the data in the band corresponding to the quantization unit 1 to the quantization unit 12 is reproduced.
[0247]
Note that whether or not to refer to the normalization coefficient of the quantization unit 13 and the quantization unit 14 included in the control data of the trial frame is other than the version matching of the data reproduction device 81 or the input of a password. For example, it may be determined in advance by the setting of the data reproducing device 81. Further, when high-quality sound reproduction is performed, the code string restoration unit 93 can restore the normalization coefficients of all the quantization units based on the additional data supplied from the control unit 92, so that the original data Needless to say, will be played.
[0248]
Next, a description will be given of a process in which the trial listening data is recorded on the recording medium, or the original data is restored from the trial listening data and the additional frame and recorded on the recording medium.
[0249]
FIG. 29 is a block diagram showing the configuration of the data recording device 141.
[0250]
Note that portions corresponding to those in the case of the data reproducing device 81 in FIG. 22 are denoted with the same reference numerals, and description thereof will be omitted as appropriate.
[0251]
That is, the code string decomposing unit 91 receives input of the encoded trial listening data, decomposes the code string and extracts the code of each signal component, and the control unit 92 receives the user's input from the operation input unit (not shown). In response to the operation, the input data is recorded with high sound quality, that is, information indicating whether or not to execute the process of restoring and recording the original data is received, and additional data is input, and the code string The processing of the restoration unit 93 is controlled.
[0252]
Based on the control of the control unit 92, the code string restoration unit 93 outputs the inputted encoded frame as it is to the recording unit 151, and restores and records the original data. In this case, the input audition data is encoded from the original data based on various information such as the audition area information, the true encoding coefficient information, and the true spectrum coefficient information supplied from the control unit 92. A process of restoring to a frame is executed, and the encoded frame of the restored original data is output to the recording unit 151.
[0253]
The recording unit 151 records data by a predetermined method on a recording medium such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or a magnetic tape. Further, the recording unit 151 may record information therein, such as a memory provided on a substrate or a hard disk. For example, when the recording unit 151 can record data on an optical disc, the recording unit 151 includes an encoder that converts data into a format suitable for recording on the optical disc, a laser light source such as a laser diode, and various lenses. And an optical unit composed of a deflection beam splitter, a spindle motor that rotationally drives the optical disc, a drive unit that drives the optical unit to a predetermined track position of the optical disc, and a control unit that controls them.
[0254]
Note that the recording medium mounted on the recording unit 151 may be the same as the recording medium on which the audition data input to the code string decomposition unit 91 or the control unit 92 or additional data is recorded. good.
[0255]
Next, data recording processing executed by the data recording device 141 will be described with reference to the flowchart of FIG.
[0256]
In step S 81, the code string decomposing unit 91 receives an input of the encoded frame of the trial listening data. In step S 82, the code string decomposing unit 91 decomposes the input code string and outputs it to the code string restoring unit 93.
[0257]
In step S83, the code string restoration unit 93 determines whether or not high sound quality recording is performed based on the signal input from the control unit 92, and if it is determined that high sound quality recording is performed, In step S84, the code string restoration process described with reference to the flowchart of FIG. 28 is executed.
[0258]
If it is not determined in step S83 that high sound quality recording is to be performed, or after the processing in step S84 is completed, in step S85, the recording unit 151 stores the code string corresponding to the input original data or trial data. Is recorded on the attached recording medium or the like.
[0259]
In step S86, the control unit 92 determines whether or not what is recorded by the recording unit 151 in step S85 is the last frame of the code string corresponding to the original data or the trial listening data.
[0260]
If it is determined in step S86 that the frame is not the last frame, the process returns to step S81, and the subsequent processes are repeated. If it is determined in step S86 that the frame is the last frame, the process ends.
[0261]
By applying the present invention, the value of the normalization coefficient information is changed, and further, the data (for example, the corresponding spectral coefficient information) that is not referred to during the reproduction process is controlled by changing the value of the normalization coefficient information. Audition data can be generated in place of information. It is very difficult to infer the original data from such sampled data, and if the sampled data is illegally altered, it may cause the sound quality to deteriorate. It is possible to protect the rights.
[0262]
When the sample data is reproduced, even if a plurality of sample areas that can be sampled with low sound quality are set in addition to one place, a part of the sample frame constituting the sample data not related to the reproduction process (for example, , A region corresponding to spectrum coefficient information that is not referred to) is inserted, and control information indicating whether or not the corresponding sample listening frame can be sampled is referred to, so that only the sample listening frame corresponding to the sample listening region is selectively selected. To be played. Accordingly, when the trial listening data is reproduced, the trial listening area is reproduced immediately after the reproduction is started, and unnatural silence reproduction is not performed.
[0263]
Further, by describing information such as a trial listening start flag and a trial listening end flag in the control information, for example, when there are a plurality of trial listening areas, fade-in and fade-out are performed at the respective start and end positions when reproducing the trial data. Thus, the audition area that should not be continuous is not reproduced continuously at the same volume, and it is possible to provide audition data that is easy to hear for the user.
[0264]
Furthermore, a part of the true value (for example, true normalization coefficient information, true spectrum coefficient information, true quantization accuracy information, or the number of quantization units) corresponding to the replaced data is included in the control information. Or by describing information such as a reproducible flag or the like, for example, depending on the type or version of the playback device, or whether there is an input of a predetermined keyword, etc. It becomes possible to control the quality.
[0265]
Then, a true value corresponding to the data changed or replaced at the time of generating the audition data (for example, true normalization coefficient information, true spectrum coefficient information, true quantization accuracy information, or quantization unit) Since the additional data composed of the additional frames in which the number is described is created, it is possible to restore the original data from the trial listening data using the additional data.
[0266]
Furthermore, when part or all of the information described in the additional frame is included in the control information, the data capacity of the additional data can be reduced. For example, the user downloads the additional data. The communication time can be shortened, for example, when trying to obtain it.
[0267]
By applying the present invention, the trial listening data and the restored original data can be reproduced and output, or can be output to other devices via a network or the like recorded on a recording medium.
[0268]
In the above, the process of generating the audition data of the content data by the audio signal and the corresponding additional data, restoring the original data from the audition data and the additional data, reproducing, and recording has been described. It is also possible to adapt to content data including an image signal or an image signal and an audio signal.
[0269]
For example, when content data by an image signal is converted using a two-dimensional DCT and quantized using various quantization tables, a dummy quantization table is designated with a high-frequency component missing, If necessary, control information is recorded in the region of the spectral coefficient information in the high frequency part corresponding to the dummy to obtain audition data. In the additional data, the missing high frequency component of the quantization table and the replaced spectral coefficient information are described.
[0270]
At the time of restoring the original data, the true quantization table in which the high frequency component is not lost is restored using the additional data, and the true spectral coefficient information is restored. Therefore, the original data is restored and decoded. be able to.
[0271]
The series of processes described above can be executed by hardware, but can also be executed by software. In this case, for example, the encoding device 1, the data reproducing device 81, or the data recording device 141 is configured by a personal computer 161 as shown in FIG.
[0272]
In FIG. 31, the CPU 171 executes various processes in accordance with a program stored in the ROM 172 or a program loaded from the storage unit 178 to the RAM 173. The RAM 173 also appropriately stores data necessary for the CPU 171 to execute various processes.
[0273]
The CPU 171, ROM 172, and RAM 173 are connected to each other via a bus 174. An input / output interface 175 is also connected to the bus 174.
[0274]
The input / output interface 175 includes an input unit 176 including a keyboard and a mouse, an output unit 177 including a display and a speaker, a storage unit 178 including a hard disk, and a communication unit 179 including a modem and a terminal adapter. It is connected. The communication unit 179 performs communication processing via a network including the Internet.
[0275]
A drive 180 is connected to the input / output interface 175 as necessary, and a magnetic disk 191, an optical disk 192, a magneto-optical disk 193, a semiconductor memory 194, or the like is appropriately mounted, and a computer program read from them is loaded. It is installed in the storage unit 178 as necessary.
[0276]
When a series of processing is executed by software, a program constituting the software executes various functions by installing a computer incorporated in dedicated hardware or various programs. For example, a general-purpose personal computer is installed from a network or a recording medium.
[0277]
As shown in FIG. 31, this recording medium is distributed to supply a program to a user separately from the apparatus main body, and includes a magnetic disk 191 (including a floppy disk) on which a program is stored, an optical disk 192 ( CD-ROM (including Compact Disk-Read Only Memory), DVD (including Digital Versatile Disk), magneto-optical disk 193 (including MD (Mini-Disk) (trademark)), or package media including semiconductor memory 194 In addition to being configured, it is configured by a ROM 172 storing a program, a hard disk included in the storage unit 178, and the like supplied to the user in a state of being incorporated in the apparatus main body in advance.
[0278]
In the present specification, the step of describing the program stored in the recording medium is not limited to the processing performed in chronological order in the order in which it is included, but is not necessarily processed in chronological order, either in parallel or individually. The process to be executed is also included.
[0279]
As mentioned above, although embodiment of this invention was described, this invention is not limited to this structure. In other words, it is understood that the invention described in the scope of claims and equivalent inventions are included in the scope of patent rights to be given to the present invention.
[0280]
For example, in the above-described embodiment, the first data string having the encoding format illustrated in FIG. 8 is illustrated, but the present invention is not limited to such a configuration. In the present invention, various other encoding formats can be applied to the first data sequence as long as the encoding format can generate the second data sequence. Here, the second data string is data that is generated by replacing the first data in the first data string with the second data, and can be reproduced in substantially the same manner as the first data string. Is a column. In the present invention, as an encoding format of the first data string, for example, at least a part (for example, a spectrum coefficient, a pixel value, etc.) related to content data on the code string and a code necessary for decoding the code It is preferable to apply a format in which quantization parameters (for example, quantization accuracy information, normalization coefficient information, etc.) are multiplexed. In this case, for example, meta information describing copyright attributes, copyright management information, encryption information, or the like can be multiplexed on the code string.
[0281]
【The invention's effect】
As described above, according to the first aspect of the present invention, a data string can be converted.
Further, according to the first aspect of the present invention, for example, only a part of an original data string such as original data can be reproduced, and a part that cannot be reproduced is skipped without being processed during the reproduction process. For example, it is possible to generate a data string such as additional data necessary for conversion to a data string such as trial listening data including control information and to restore the original data string from the converted data string.
[0282]
According to the second aspect of the present invention, it is possible to receive and reproduce a data string.
Further, according to the second aspect of the present invention, for example, silence can be obtained by determining a reproducible part and a non-reproducible part based on the control information and continuously reproducing only the reproducible part. Or reproduction of only noise can be avoided.
[0283]
According to the third aspect of the present invention, the data string can be restored.
In addition, according to the third aspect of the present invention, for example, by acquiring a data string such as additional data for restoring the data string separately, it becomes a basis of original data or the like from the data string such as the audition data. The data column can be restored.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of an encoding apparatus to which the present invention is applied.
FIG. 2 is a block diagram illustrating a configuration of a conversion unit in FIG. 1;
FIG. 3 is a diagram illustrating a spectrum signal and a quantization unit.
4 is a block diagram showing a configuration of a signal component encoding unit in FIG. 1. FIG.
FIG. 5 is a diagram for explaining a tone component and a non-tone component.
6 is a block diagram showing a configuration of a tone component encoding unit in FIG. 4;
7 is a block diagram showing a configuration of a non-tone component encoding unit in FIG. 4. FIG.
FIG. 8 is a diagram for explaining a format of a frame of original data.
9 is a block diagram showing a configuration of a trial listening data generation unit of FIG. 1. FIG.
FIG. 10 is a diagram illustrating a trial listening area and a protection area of an input frame sequence.
FIG. 11 is a diagram for explaining a format of a trial listening frame.
12 is a diagram illustrating a spectrum signal corresponding to the trial listening frame of FIG. 11;
FIG. 13 is a diagram for explaining a trial listening frame in which a part of a spectrum coefficient is rewritten using control information.
FIG. 14 is a diagram illustrating control information.
FIG. 15 is a diagram illustrating an additional frame.
FIG. 16 is a diagram illustrating control information.
FIG. 17 is a flowchart for explaining trial listening data generation processing;
FIG. 18 is a flowchart for explaining trial listening data generation processing;
FIG. 19 is a diagram illustrating a frame of original data when tone components are not separated.
FIG. 20 is a diagram for explaining a trial listening frame when tone components are not separated.
FIG. 21 is a diagram illustrating an additional frame when tone components are not separated.
FIG. 22 is a block diagram showing a configuration of a data reproducing apparatus to which the present invention is applied.
23 is a block diagram showing a configuration of a signal component decoding unit in FIG.
24 is a block diagram showing a configuration of a tone component decoding unit in FIG. 23. FIG.
25 is a block diagram showing a configuration of a non-tone component decoding unit in FIG. 23. FIG.
26 is a block diagram showing a configuration of an inverse conversion unit in FIG.
FIG. 27 is a flowchart for describing data reproduction processing;
FIG. 28 is a flowchart illustrating a code string restoration process.
FIG. 29 is a block diagram showing a configuration of a data recording apparatus to which the present invention is applied.
FIG. 30 is a flowchart for describing data recording processing;
FIG. 31 is a block diagram illustrating a configuration of a personal computer.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 Encoding apparatus, 11 Conversion part, 12 Signal component encoding part, 13 Code sequence generation part, 14 Audition data generation part, 61 Audition data generation process control part, 62 Audition area determination part, 63 Band-limiting process part, 64 Control Information insertion unit, 65 additional frame generation unit, 66 audition data generation unit, 67 additional data generation unit, 92 control unit, 93 code string restoration unit

Claims (25)

In the data conversion method of the data conversion device for converting the first data string for reproducing the predetermined information into the second data string for reproducing the predetermined information by degrading the reproduction quality of the predetermined information ,
When the first data included in the first data string and used for reproducing the predetermined information is used for reproducing the predetermined information. A first replacement step of replacing the reproduction quality of the predetermined information with second data that deteriorates compared to the case of the first data sequence ;
Wherein the first data included in the different regions, at least a portion of the third data is at least a portion of the data used to reproduce the predetermined information, the second data The second data is replaced with fourth data which is control information to be referred to when the column is reproduced , and the other part is a non-reference area which is not referred to as reproduction data when the second data string is reproduced . A replacement step of
A first generation step of generating the second data string using the data generated by the processing of the first replacement step and the processing of the second replacement step;
Data conversion method, which comprises a. Each of the first data string and the second data string is composed of a plurality of frames,
The third data is recorded in each of the plurality of frames,
The data conversion method according to claim 1, wherein the fourth data is control information indicating whether or not a corresponding frame can be reproduced. A second generation step of generating a third data string necessary for restoring the second data string generated by the processing of the first generation step into the first data string;
The third data string generated by the process of the second generation step includes the first data and is converted into the fourth data by the process of the second replacement step among the third data. The data conversion method according to claim 1, further comprising a replaced part . A frequency component conversion step for converting input data into frequency components;
An encoding step of encoding the data converted into frequency components by the processing of the frequency component conversion step;
In the process of the first replacement step, the encoded data encoded by the process of the encoding step is replaced with the first data string, and the first data is replaced with the second data.
In the process of the second replacement step, the encoded data encoded by the process of the encoding step is the first data string, and at least a part of the third data is the second data. data conversion method according to claim 1, characterized in that replace. The third data includes at least a part of spectrum coefficient information of the frequency component converted by the processing of the frequency component conversion step.
  The data conversion method according to claim 4, wherein: The first data includes at least a part of normalization coefficient information of an encoding process by the process of the encoding step.
  The data conversion method according to claim 4, wherein: The first data includes at least a part of quantization accuracy information of an encoding process by the process of the encoding step.
  The data conversion method according to claim 4, wherein: The first data is encoded by the encoding step. Includes at least some of the information representing the number of quantization units
  The data conversion method according to claim 4, wherein: The first data is data indicating a predetermined numerical value,
  The second data is obtained by minimizing the numerical value of the first data.
  The data conversion method according to claim 1. The second data is obtained by replacing at least a part of the first data with random data.
  The data conversion method according to claim 1. In the data conversion device for converting the first data string for reproducing the predetermined information into the second data string for reproducing the predetermined information with degraded reproduction quality ,
When the first data included in the first data string and used to reproduce the predetermined information is used to reproduce the predetermined information. First replacement means for replacing the reproduction quality of the predetermined information with second data that deteriorates compared to the case of the first data string ;
Wherein the first data included in the different regions, at least a portion of the third data is at least a portion of the data used to reproduce the predetermined information, the second data The second data is replaced with the fourth data which is control information referred to when the column is reproduced , and the other part is a non-reference area which is not referred to as reproduction data when the second data string is reproduced . Replacement means,
Generating means for generating the second data string using data generated by the first replacing means and the second replacing means;
Data conversion apparatus comprising: a. A computer-executable program for controlling a data conversion apparatus for converting a first data string for reproducing predetermined information into a second data string for reproducing the predetermined information with degraded reproduction quality Because
When the first data included in the first data string and used for reproducing the predetermined information is used for reproducing the predetermined information. A first replacement step of replacing the reproduction quality of the predetermined information with second data that deteriorates compared to the case of the first data sequence ;
Wherein the first data included in the different regions, at least a portion of the third data is at least a portion of the data used to reproduce the predetermined information, the second data The second data is replaced with fourth data which is control information to be referred to when the column is reproduced , and the other part is a non-reference area which is not referred to as reproduction data when the second data string is reproduced . A replacement step of
A generation step for generating the second data string using the data generated by the processing of the first replacement step and the processing of the second replacement step;
The program characterized by including . Generated on the basis of the first data string for reproducing the predetermined information, the data reproducing method of the data reproducing apparatus for reproducing second data string for reproducing deteriorated reproduction quality of the predetermined information In
The first data string includes first data which is at least a part of data used for reproducing the predetermined information, and is in a different area from the first data. Third data that is included and is at least part of data used to reproduce the predetermined information is further included.
The second data string indicates the reproduction quality of the predetermined information when the first data in the first data string is used for reproducing the predetermined information. Control information that is replaced with second data that deteriorates more than in the case of the above, and that is referred to when at least a part of the third data in the first data string reproduces the second data string And the other portion is a non-reference area that is not referred to as reproduction data when the second data string is reproduced ,
A data reproduction method comprising: a reproduction control step for controlling reproduction of the second data string based on the fourth data included in the second data string. The fourth data includes information specifying whether or not a corresponding frame can be reproduced when reproduction of the second data string is controlled by the process of the reproduction control step. The data reproduction method according to claim 13 . The data reproduction method according to claim 13 , wherein the fourth data includes at least a part of the first data replaced with the second data. For the fourth data , when reproduction of the second data string is controlled by the process of the reproduction control step, at least a part of the first data included in the control information is used. The data reproducing method according to claim 15 , further comprising information indicating whether or not the second data can be restored. An inverse transformation step for inversely transforming the frequency component of the second data string ;
Data reproducing method according to claim 13, wherein <br/> further comprising a decoding step of decoding the inverse transformed second data string by the processing of the inverse transform step. The third data includes at least a part of spectrum coefficient information of the frequency component converted by the processing of the frequency component conversion step.
  The data reproducing method according to claim 17, wherein: The first data includes at least a part of normalization coefficient information of an encoding process by the process of the encoding step.
  The data reproducing method according to claim 17, wherein: The first data includes at least a part of quantization accuracy information of an encoding process by the process of the encoding step.
  The data reproducing method according to claim 17, wherein: The first data includes at least a part of information indicating the number of quantization units of the encoding process by the process of the encoding step.
  The data reproducing method according to claim 17, wherein: The first data is data indicating a predetermined numerical value,
The data reproduction method according to claim 13 , wherein the second data is obtained by minimizing a numerical value of the first data. The second data is obtained by replacing at least a part of the first data with random data.
  The data reproduction method according to claim 13, wherein: A computer for controlling a data reproducing apparatus that reproduces a second data string that is generated based on a first data string for reproducing predetermined information and that reproduces the predetermined information with degraded reproduction quality. Is an executable program,
The first data string includes first data which is at least a part of data used for reproducing the predetermined information, and is in a different area from the first data. Third data that is included and is at least part of data used to reproduce the predetermined information is further included.
The second data string indicates the reproduction quality of the predetermined information when the first data in the first data string is used for reproducing the predetermined information. Control information that is replaced with second data that deteriorates more than in the case of the above, and that is referred to when at least a part of the third data in the first data string reproduces the second data string And the other part is a non-reference area that is not referred to as reproduction data when the second data string is reproduced.
sometimes,
Wherein based on said fourth data contained in the second data string, the program characterized by comprising a reproduction control step for controlling reproduction of the second data string. A second data string that is generated based on the first data string for reproducing the predetermined information and is reproduced with the reproduction quality of the predetermined information deteriorated is restored to the first data string. In the data restoration method of the data restoration device to
The first data string includes first data that is at least a part of data used for reproducing the predetermined information;
The second data string indicates the reproduction quality of the predetermined information when the first data in the first data string is used for reproducing the predetermined information. Is replaced with second data that deteriorates more than in the case of
An acquisition control step for controlling acquisition of a third data string including information necessary to restore the second data string to the first data string;
A restoration step of restoring the first data string from the second data string based on the third data string whose acquisition is controlled by the processing of the acquisition control step;
Said third data string comprises the first data for restoring the first data string, wherein in the process of the restoration step, the acquisition by the processing of the control step acquisition controlled the third It said first data contained in the data string, by replacing the second data contained in the second data stream, restores the first data row,
A part of said 2nd data is control information referred when reproducing said 2nd data sequence. The data restoration method characterized by the above-mentioned.

Images