JP2003249026A - Reproducing device and reproducing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To realize the varying-speed reproduction faster than the speed at the recording operation without using a high speed decoder circuit. <P>SOLUTION: Frames are read from an HDD at the data speed of m/n times (m, n are integers, and m/n>1) the speed at the recording operation. From the read frames, the information showing the sound volume is taken out for every frame. By the information showing the taken out sound volume, n pieces of frames provided with digital audio data having the large sound volume among m pieces of continuously recorded frames are preferentially taken out. The taken out n pieces of frames are continuously supplied to the decoder circuit 413, then the digital audio data are obtained at the speed of m/n times the speed at the recording operation. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a reproducing device and a reproducing method.

[0002]

2. Description of the Related Art Video signals and audio signals are transmitted, for example, to MP.
A technique of compressing data by the EG method, recording it in the HDD, and reproducing it has been put to practical use. However, when reproducing a video signal and an audio signal that have been compressed by data and recorded at a speed higher than the recording speed, generally, only the video is displayed and the audio is not output. In the following description, the process or function of reproducing at a speed higher than the recording speed will be referred to as "variable speed reproduction".

However, when video and audio are recorded,
Audio often has important meaning, for example, when searching for a specific recording position by variable speed playback, if you can hear not only the image but also the audio, you can search for the target recording position by using the audio as a clue, Easy to find Therefore, it is preferable that the voice can be reproduced during the variable speed reproduction.

Therefore, as a method of reproducing sound during variable speed reproduction, methods shown in FIGS. 7 and 8 have been considered. 7 and 8 show the case of variable speed reproduction at 2 × speed. In the method of FIG. 7, as shown in FIGS. 7A and 7B, compressed data is extracted from the HDD at 2 × speed at the time of recording, and this is It is supplied to the decoder circuit. Note that reference numerals 1, 2, ..., 2N are numbers attached to data blocks, for example, data frames for each unit period.

Then, in the decoder circuit, as shown in FIG. 7C, the supplied data is decoded at a double speed to output audio data at a double speed, and this audio data is written in the buffer memory. And
As shown in FIG. 7D, the written audio data is thinned out every other data frame, and 1 /
It is read at a speed of 2. Therefore, it is possible to obtain audio data of double speed reproduction from the buffer memory.

In addition, in the method of FIG.
As shown in FIG. 8, the compressed data extracted at the double speed at the time of recording is directly written to the buffer memory, and as shown in FIG. The read data is supplied to the decoder circuit and decoded. Therefore, as shown in FIG. 8D, audio data for double speed reproduction can be obtained.

[0007]

However, in the case of the method of FIG. 7, a decoder circuit capable of performing a decoding process at double speed is required, but it is technically difficult to realize this at present. is there. On the other hand, in the case of the method of FIG. 8, since the decoder circuit only needs to perform the decoding process at 1 × speed, that is, the decoding speed becomes the same as that at the time of normal reproduction, an IC for high-speed processing is required and it is expensive It never becomes.

However, the method of FIG. 7 is the same as that of FIG.
In the case of method, data is thinned out at regular intervals, so
For example, in the case of speech or the like, an important part for understanding the content may be thinned out, and the understanding level of the content may decrease. Also, when performing variable-speed reproduction of musical tones, data is thinned out without considering the characteristics of the signal, so it may be difficult to hear, such as the tempo shifting depending on the content.

Further, regardless of the method shown in FIG. 7 or the method shown in FIG. 8, simply thinning out the data will result in continuation of data that was not continuous at the time of recording, so that the waveforms cannot be smoothly connected. However, noise is generated at the connection point and the quality deteriorates.

The present invention is intended to solve the above problems.

[0011]

According to the present invention, for example, digital audio data is made into one frame for each predetermined number of samples, data compression is performed for each frame, and reproduction of digital audio data in each frame is performed. The above-mentioned frame is reproduced from the medium in which the information indicating the volume at each time is recorded by being integrated with the corresponding frame for each corresponding frame, and the data compression of the reproduced frame is complementary to the above-mentioned data compression. When the digital audio data is reproduced by performing the processing of step m, the m /
Play at a speed n times (m and n are integers, m / n> 1),
Information indicating the volume is extracted from the reproduced frame for each frame, and digital audio data having a high volume is extracted from the continuously recorded m frames by the extracted information indicating the volume. The n frames that are included are preferentially taken out, and the n frames are continuously supplied to the decoder circuit so that the digital audio data can be obtained at m / n times speed at the time of recording.

[0012]

FIG. 1 shows an example of the present invention,
In this example, a video signal can be recorded and reproduced, and an HDD is used as a recording medium and a recording device. Also, in the following example, MP is used as the data compression / decompression method for video signals and audio signals.
This is a case where the EG method is adopted. In FIG. 1, circuits 11 to 26 are recording systems, circuits 31 to 33 are recording / reproducing common parts, circuits 41 to 55 are reproducing systems, and a circuit 60 is a system controller for controlling the operation of the entire apparatus.

At the time of recording, the tuner 11 receives the television broadcast to extract the color composite video signal and the audio signal, and these signals are supplied to the input selector circuit 14. In addition, external color composite video signals and audio signals are input to the input terminals 12V and 12A.
Is supplied to the input selector circuit 14 through. further,
External luminance signal and carrier color signal (so-called S video signal) are input switching circuit 22 through input terminal 13V.
The audio signal paired with the luminance signal and the carrier color signal is also supplied to the input selector circuit 14 through the input terminal 13A.

The input selector circuit 14 selects the video signal and the audio signal supplied thereto, and the audio signal of the selected video signal and audio signal is supplied to the A / D converter circuit 15. A / D conversion is performed on the digital audio data, and the digital audio data is supplied to the MPEG encoder circuit 16 and compressed into MPEG layer 2 system data.

In the MPEG layer 2 system, a 1-channel audio signal (digital audio data) is band-divided into 32 sub-bands, and 36 samples of each sub-band form one set (= 32 sub-bands × 36). Sample = 1152 samples), and one set is called "frame".

Then, a specific data compression process (encoding process) is executed for each frame, and the frame after the data compression is the actually compressed data of the voice data,
And the parameter at the time of compression. In this case, the audio data is normalized for each sub-band, and the audio data of the sub-band that can be ignored in terms of hearing is removed. In addition, the parameter at the time of compression includes an exponent part at the time of normalization of audio data, and this exponent part is called a “scale factor”. Therefore, this scale factor is also information indicating the signal level of each subband, that is, the volume level.

In this way, the MPEG encoder circuit 16 generates an audio elementary stream, and this elementary stream is supplied to the multiplex / demultiplex circuit 31.

Further, from the input selector circuit 14, the A / D
The component video signal paired with the audio signal supplied to the converter circuit 15 is taken out, and this signal is Y / C.
It is supplied to the separation circuit 21 and separated into a luminance signal and a carrier color signal, and these signals are supplied to the input switching circuit 22. The input switching circuit 22 is controlled by the system controller 60 in conjunction with the input selector circuit 14, and when the input selector circuit 14 selects the output signal of the tuner 11 or the input terminals 12V and 12A, The luminance signal and the carrier color signal output from the Y / C separation circuit 21 are output, and the input selector circuit 14
However, when the output signal of the input terminal 13A is selected, the luminance signal and the carrier color signal obtained from the input terminal 13V are output.

Therefore, the input switching circuit 22 outputs the luminance signal and the carrier color signal in the video signal supplied through the tuner 11 or the input terminals 12V and 13V. Then, the luminance signal and the carrier color signal are supplied to the NTSC decoder circuit 23, decoded into component video signals, and A / D converted into digital image data, and the digital image data is supplied to the pre-video signal processing circuit 25. It is supplied to the MPEG encoder 26 after being subjected to various kinds of video signal processing such as a pre-filter.

At this time, the NTSC decoder circuit 2
A vertical synchronizing signal, a horizontal synchronizing signal and a burst signal are supplied from 3 to the synchronizing control circuit 24 to generate clocks and synchronizing signals of various timings necessary for recording processing, and these signals are supplied to the respective circuits.

In the MPEG encoder circuit 26, the digital image data supplied thereto is subjected to data compression by encoding processing such as block DCT to generate an elementary stream of the image, and the elementary stream of the image is generated. Are supplied to the multiplex / demultiplex circuit 31.

During recording, the multiplex / demultiplex circuit 31 multiplexes the audio elementary stream supplied thereto, the image elementary stream, and various control signals, and transports them in the MPEG system. When a stream is generated and reproduced, a process complementary to that at the time of recording is performed.

When recording, multiplex /
The transport stream generated in the demultiplex circuit 31 is passed through the buffer circuit 32 to HD.
It is supplied to D33. In this case, the buffer circuit 32
This is for intermittently supplying a transport stream continuously input to the HDD 33.
That is, when the HDD 33 is performing a seek operation, data cannot be written, but the transport stream is continuously output from the multiplex / demultiplex circuit 31, or the HDD 3 is used.
It is output regardless of the seek period of 3.

Therefore, the buffer circuit 32 is connected to the HDD 33.
When writing is not possible, the transport stream output from the multiplex / demultiplex circuit 31 is temporarily stored, and when it is writable, the stored transport stream is stored.
The data is output to the HDD 33 at a rate higher than the rate at which it was output. Therefore, the transport stream output from the multiplex / demultiplex circuit 31 is written in the HDD 33 without any loss.

The HDD 33 is controlled by the system controller 60 and records the transport stream at a predetermined address during recording. IDE can be used as an interface between the buffer circuit 32 and the HDD 33.

As described above, the video signal and the audio signal of the tuner 11 and the input terminals 12V and 12A or 13V and 13A are recorded in the HDD 33.

The system controller 60 controls the circuits so that the circuits execute various processes required for recording and reproducing in accordance with a user's instruction. Therefore, the system controller 60 is composed of, for example, a microcomputer, and has a CPU 61, a ROM 62 in which a control program and the like are written, a work area R, and the like.
It has an AM 63, an interface circuit 64 for a user interface, and an interface circuit 65 for accessing control signals and data to each circuit.

On the other hand, during normal reproduction, the HDD 33 is controlled by the system controller 60 to seek to a predetermined address, the transport stream is read, and the read transport stream is made continuous in the buffer circuit 32. It is supplied to the multiplex / demultiplex circuit 31 after being processed.

Then, in the multiplex / demultiplex circuit 31, PES is extracted from the transport stream supplied thereto, and this PES is M
It is supplied to the PEG-AV decoder circuit 41, the video elementary stream and the audio elementary stream are separated, and each is subjected to data expansion (MPEG decoding processing) to output baseband digital image data and digital audio data. To be done.

Then, the digital audio data is supplied to the D / A converter circuit 43 through the switching circuit 42 and D / A converted into the original audio signal, and this audio signal is taken out to the output terminal 44. In addition, MPEG-
The digital image data output from the AV decoder circuit 41 is supplied to the post video signal processing circuit 51 to be subjected to filter processing and the like, and then supplied to the OSD circuit 52. The OSD circuit 52 generates graphics for screen display and synthesizes it with digital image data,
This is for processing the displayed image.

Then, the digital image data output from the OSD circuit 52 is supplied to the NTSC encoder circuit 53, where it is subjected to processing and D / A conversion reverse to those of the NTSC circuit 23, and an analog color composite video signal,
A luminance signal and a carrier color signal (S video signal) are formed, and these signals are output to the output terminals 54 and 55.

As described above, the video signal and the audio signal recorded in the HDD 33 are reproduced, and the output terminal 4
It is output to 4, 54 and 55.

The output signal of the A / D converter circuit 15 is passed through the switching circuit 42 to the D / A converter circuit 4
3 and the digital image data output from the pre-video signal processing circuit 25 and the digital image data obtained by decoding the signal recorded in the HDD 33 by the MPEG-AV decoder circuit 41 are post-video signal processed. By supplying both digital image data to the circuit 51 and mixing or synthesizing them, an image signal obtained by mixing or synthesizing the image reproduced by the HDD 33 with the image obtained by the tuner 11 or the input terminal 12V or 13V can be obtained.

Recording and normal reproduction of the video signal and the audio signal are realized as described above. However, in order to enable variable speed reproduction of the audio signal, the present invention is further configured as follows. .

In the following description, the variable reproduction speed of the audio signal is expressed as m / n times the recording speed (values m and n will be described later). Therefore, m / n> 1, but in practice, the maximum value of the variable reproduction speed m / n is about several times. Further, in the following example, as a numerical example, m / n = 2, that is, a case of double speed reproduction is shown.

During this m / n double speed reproduction, HD
The data recorded in D33 is read out at an average speed of m / n times that at the time of recording.
The PES is continuously output from the demultiplex circuit 31 at a speed of m / n times as high as that at the time of recording.
The ES is supplied to the MPEG-AV decoder circuit 41.

The MPEG-AV decoder circuit 41 is
The audio decoder unit is configured as shown in FIG. 2, for example. That is, in FIG. 2, reference numeral 41A indicates MPEG.
The audio decoder section of the AV decoder circuit 41 is shown, and at the time of variable speed reproduction, the multiplex / demultiplex circuit 3
1 is supplied to the separation circuit 411 of the MPEG-AV decoder circuit 41 to separate the video elementary stream and the audio elementary stream, and the audio elementary stream is buffered by the buffer circuit 4.
12 and the frames in the audio elementary stream are written in the frame order as shown in FIG. 3A.

FIG. 3 schematically shows the audio elementary stream in frame units, and the vertical axis represents the analog signal level when the frame is decoded and D / A converted, that is, included in each frame. 2 shows the volume level indicated by the scale factor. Therefore, the height of each frame indicates the volume level of the audio signal obtained by decoding and D / A converting the frame. Further, at this time, the data rate of the frames written in the buffer circuit 412 is double. Further, reference numerals 1, 2, ... Are numbers attached to the frames.

Then, m consecutive to the buffer circuit 412
After the writing of the frames, the m frames are supplied to the parameter extraction circuit 414, and the volume information in each frame, in this case, the scale factor (a total of m frames) is extracted and extracted. The other scale factors are supplied to the discrimination circuit 415, and among the m scale factors, the top n scale factors showing a large audio signal level are discriminated.

Here, when the value m is small, the effect of the present invention may be reduced, as will be apparent from the later description. Therefore, it is preferable to increase the value m. However, if the value m is set too large, the capacity of the buffer circuit 412 needs to be increased, and the processing time may become long. Therefore, in consideration of the above, for example, m = 90. Further, since the value m / n becomes the speed of variable speed reproduction as described above, in the case of double speed reproduction, if m = 90, then n = 45. The setting of the numerical values m and n and the control of the overall operation of the circuit of FIG.
This is performed by the system controller 60.

Then, the discrimination result of the discrimination circuit 415 is supplied to the buffer circuit 412, and as shown in FIG. 3B, among the m frames, the frames corresponding to the upper n scale factors having a high signal level are valid. The frames corresponding to the remaining (mn) scale factors (frames marked with an X in FIG. 3A) are thinned out. That is, by checking the scale factor, among the m consecutive frames, the upper n frames with the higher audio signal level are left (in the valid state), and the lower (m−n) frames with the lower audio level. Is deleted (invalid state).

Then, as shown in FIG. 3C, the upper n frames left in the buffer circuit 412 are read from the buffer circuit 412 at a data rate n / m times as high as that at the time of writing, and read. The frame is supplied to the decoder circuit main body (decoder core unit) 413 and is decompressed into the original digital audio data. The decompressed digital audio data is supplied to the D / A converter circuit 43 through the switching circuit 42,
After that, the processing is performed as described above. The above operation is repeated for every m frames.

In this case, the number of frames output from the buffer circuit 412 is thinned out to n / m, and
Since the data is read at a data rate n / m times as high as that at the time of writing, the data rate of the read frame is 1 ×. Further, since the data read from the HDD 33 at m / n speed (for example, 2 speed) is thinned out to n / m times, the data speed of the frame output from the buffer circuit 412 is 1 time speed and at the same time, Are taken out continuously. Therefore, the audio signal by the double speed reproduction can be continuously obtained at the terminal 44.

In this way, according to the above-mentioned apparatus, it is possible to carry out variable speed reproduction at m / n times speed.
Since the data speed of the frame supplied to the decoder circuit body 413 of the G-AV decoder circuit 41 is 1 × speed, that is, the decoding speed is the same as that in normal reproduction, as in the case of FIG. It does not require a processing IC and is not expensive.

Furthermore, when thinning out the data read out from the HDD 33 at m / n times speed, a frame having a large volume level is left out of the read out data, so that when a speech, a drama or a movie dialogue is used. , It is possible to reduce the thinning out of the important parts in understanding the contents, and to suppress the deterioration of the understanding level of the contents.

By the way, as described above, when variable speed reproduction is realized by connecting only frames with a high volume level,
Since the frames that were not continuous at the time of recording are made continuous, the waveform is not smoothly connected at the connection point, so noise may occur and the quality may deteriorate.

In order to solve such a problem, the audio decoder section 41A shown in FIG. 2 is further provided with a rewrite control circuit 416. Then, a part of the discrimination output of the discrimination circuit 415 is supplied to the rewrite control circuit 416 to form a predetermined control signal, and this control signal is supplied to the buffer circuit 412, which is positioned before and after the frame to be thinned out. The frame scale factor is rewritten.

FIG. 4 schematically shows a state in which the scale factor of the frame stored in the buffer circuit 412 is rewritten by the rewriting control circuit 416.
A is the same as FIG. 3B. That is, FIG. 4A shows a state in which the buffer circuit 412 deletes (m−n) frames with a low sound volume level and leaves n frames with a high sound volume level.

Then, for the following explanation, the remaining n
FIG. 4B is obtained by packing the individual frames in the time axis direction. It should be noted that the deleted frame was present between the frame before the code P and the next frame, and the code P is the connection point. And this connection point P
Then, the waveform is often discontinuous as described above.

Therefore, as shown by the solid line in FIG. 4C (the broken line is the same as in FIG. 4B), the volume level of, for example, three frames before the connection point P gradually decreases from the original level,
The scale factors included in these six frames are rewritten so that the volume of, for example, three frames subsequent to the connection point P gradually becomes the original volume from the reduced value. Note that this rewriting is performed in the buffer circuit 412 based on the control signal from the rewriting control circuit 416, as described above. In addition, the scale factors of other frames are left unchanged.

Then, the rewritten frame of FIG. 4C is read from the buffer circuit 412 at a data rate n / m times the writing rate, as described with reference to FIG. 3C. The frame is supplied to the decoder circuit body 413 and decoded into the original digital audio data.

Therefore, in the audio signal obtained by D / A converting this digital audio data, even if the waveform is not continuous at the connection point P, the connection point P
Since the volume is once low, the noise caused by the discontinuity of the waveform is not noticeable.

As described above, according to the above-described device, not only the m / n times speed variable speed reproduction can be performed, but also the noise generated during the variable speed reproduction can be reduced without deteriorating the understanding of the contents.

By the way, the audio decoder unit 4 shown in FIG.
The main part of 1A can also be configured by a DSP and software, and FIGS. 5 and 6 show an example in that case. That is, as shown in FIG.
The PES output from the demultiplex circuit 31 is
This DSP410 is supplied to the DSP410
A buffer circuit 412 is connected to.

Then, in the DSP 410, for example, the routine 100 shown in FIG. 6 is executed as follows. That is, the system controller 60 causes the DSP 410
When the variable speed reproduction is instructed to, the processing of the routine 100 starts from step 101 in the DSP 410,
Next, in step 102, the frame (one frame) of the audio elementary stream is separated from the PES which is the output of the multiplex / demultiplex circuit 31, and accumulated in the buffer circuit 412.

Then, the process proceeds to step 103, where m
It is determined whether or not the number of frames has been accumulated in the buffer circuit 412, and when the number of frames has not been accumulated, the process returns from step 103 to step 102. In this way, consecutive m frames are accumulated in the buffer circuit 412.

Then, m frames are buffer circuits 4
When the data is stored in 12, the determination is made in step 103, and the process proceeds from step 103 to step 111. In step 111, the buffer circuit 412
The scale factors of the m number of frames stored in the frame are ranked, and, for example, as shown in FIG. 4B, the upper n number of frames having a high audio signal level are left, and the (m−n) number of low volume levels are left. Frame is deleted.

Subsequently, the process proceeds to step 112. In step 112, for example, as shown in FIG. 4C, for example, three of the frames stored in the buffer circuit 412 before the connection point P are located. The scale factors are rewritten so that the frames fade out and, for example, the three frames located after the connection point P fade in.

Then, in step 113, the n frames left in the buffer circuit 412 are 1
The original digital audio data is decoded at the double frame rate one by one at a time, and the decoded digital audio data is supplied to the D / A converter circuit 43 through the switching circuit 42.

When the decoding process is completed for n frames, the process proceeds to step 114. In step 114, it is judged whether or not the end of the variable speed reproduction is instructed, and when the end is not instructed. The process returns from step 114 to step 102. Therefore, after that, the processing of steps 102 to 114 is repeated with m frames of the audio elementary stream of PES output from the multiplex / demultiplex circuit 31 as one set, so that the m / n speed is increased. It is possible to obtain a reproduction audio signal. When it is instructed to end the variable speed reproduction in step 114, the process proceeds from step 114 to step 115 to end the routine 100, and therefore,
End variable speed playback.

Therefore, even when the DSP 410 is used, when the frame whose speed has been changed is decoded into the original digital audio data, the data speed required for the decoding is 1 × speed.
There is no need for P and there is no increase in cost.

Further, among the data read from the HDD 33, the frame having a large volume level is left, so that thinning out of an important part for understanding the contents is reduced, and the deterioration of the understanding of the contents is suppressed. be able to.
Further, since the volume is low at the connection point P, noise generated during variable speed reproduction can be reduced.

In the above description, the frame compressed by the MPEG method is decoded into the original digital audio data. However, the frame compressed by the other method is decoded into the original digital audio data. Also in this case, for every m frames, the top n frames with the highest volume may be left according to the volume information for each frame. For example, A
When decoding the original digital audio data from the frame compressed by the C-3 (registered trademark) method, the exponent can be used as the volume information.

Also, when the original digital audio data is decoded from the frame compressed by the MPEG method, for example, the volume information for each channel and subband is obtained from the scale factor, and
It is also possible to weight the volume information in consideration of the minimum audible limit, average them, and leave the top n frames based on this average value. Then, by doing so, it is possible to preferentially reproduce the frame in which the human feels that the volume is actually high.

Further, it is also possible to extract only the parameter indicating the volume information of a specific sub-band such as a low frequency band and average it, and leave the top n frames based on the average value.
By doing so, the amount of calculation at the time of the decoder processing can be reduced.

When data is compressed by the AC-3 (registered trademark) method, one frame contains six audio blocks, but only the parameters of the first audio block are extracted. Even if the volume of the entire frame is determined, the volume of the frame can be determined with sufficient accuracy. Also in this case, the amount of calculation during the decoder processing can be reduced.

Further, among the m frames, the upper n frames having a large volume are validated, and the remaining (m−n) frames are thinned out. , You can skip the frame and leave the frame that should be skipped. By doing so, the number of connection points P is reduced,
The noise at the connection point P is reduced, and the reproduced sound becomes easier to hear.

On the contrary, when only a few frames to be thinned are continuous, do not thin out the frames,
You can also thin out frames that should not be thinned out. Even in such a case, the number of connection points P is reduced, so that noise is reduced and the reproduced sound is easily heard.

Further, during variable speed reproduction, the information directly input by the user, the information recorded together with the frame (digital audio data), the information obtained by analyzing the parameters included in the frame, or the like is used as described above. It is also possible to combine the process of using the top n frames with the highest volume for reproduction with the process of using the frames for reproduction at regular intervals as described in FIG. In other words, when the content of a certain source is changed to speech or music during variable speed reproduction of a certain source, during the variable speed reproduction of the speech, as described above, the upper n frames with a large volume are used for reproduction. However, during variable speed reproduction of music, frames can be used for reproduction at a constant cycle as described with reference to FIG.

In this way, since a frame with a high volume is used for reproduction during variable speed reproduction of speech, the clarity of the speech becomes high, and even during variable speed reproduction of music, even during variable speed reproduction. The tempo of the music can be kept constant, and as a result, variable-speed reproduction that is easy to hear can be performed for a wide range of contents.

Further, in that case, the contents to be reproduced at variable speed can be known by, for example, examining the band distribution of the volume level, and the tempo of the musical sound and the like are also examined with time of the volume level of the specific band. You can know it.

Further, in the above description, the decoding processing speed of the decoder circuit main body 413 and the DSP 410 is 1 × speed. However, for example, when the 3 × speed variable reproduction is performed at the 2 × processing speed, that is, the decoding is performed. The present invention can be applied to the case where variable speed reproduction is performed at a reproduction speed higher than the processing speed.

When performing fade-out / fade-in before and after the connection point P of the frame, for example, as shown in FIG. 4C, when the signal level at the connection point P is reduced, understanding of the contents when no fade processing is performed The noise level can be suppressed while maintaining the degree, but when the signal level at the connection point P is set to 0 (silence), the noise can be further suppressed.

Further, the number of frames to be faded out / in and the signal level at the connection point P can be changed according to the contents and speed of the variable speed reproduction. Alternatively, the number of frames to be faded out / in can be changed according to the number of frames between the two connection points P. In that case, the number of frames sandwiched between the two connection points P and P can be changed. When the number of frames is large, if the number of frames to be faded is increased and the fade is performed gently, it becomes easy to hear the contents reproduced at variable speed.

Depending on the compression method, the time length of one frame may be different. In that case, the fade process is performed by changing the number of frames to be faded depending on the frame length. You can adjust the time spent. Furthermore, when the data compression method of the digital audio data is the AC-3 (registered trademark) method,
The parameter rewritten when performing the fade-out / fade-in at the connection point P can be a dialnorm or the like that is always included in each frame.

Further, in the above, the buffer circuit 4
In 12, it was explained that, out of m frames, n frames are valid and (mn) frames are deleted. However, a flag should be set to the valid n frames and should be deleted. It is also possible to reset the flags of the (mn) frames, and rewrite the scale factor and supply the scale circuit from the buffer circuit 412 to the decoder circuit main body 413 for the frame in which the flags are set.

Further, in the above, as the recording medium, H
Although I used DD33,
A solid-state memory or the like can also be used.

Further, in the above, the tuner 11,
In the case of recording and reproducing analog video signals and audio signals supplied from the terminals 12V, 12A or 13V, 13A, IEEE1394 is used with an external IRD.
The present invention can also be applied to the case of recording and reproducing digital video signals and audio signals through such a digital interface.

[List of Abbreviations Used in This Specification] A / D: Analog to Digital AC-3 (registered trademark): Audio Code number 3 (Trade M
ark) CPU: Central Processing Unit dialnorm: dialog normalization D / A: Digital to Analog DCT: Discrete Cosine Transform DSP: Digital Signal Processor HDD: Hard Disk Drive IC: Integrated Circuit IDE: Integrated Drive Electronics IEEEEE: Institute of Electrical and Electronics
Engineers IRD: Integrated Receiver Decorder MPEG: Motion Picture Image Coding Experts Grou
p MPEG-AV: MPEG Audio and Video NTSC: National Television System Committee OSD: On Screen Display PES: Packetized Elementary Stream RAM: Random Access Memory ROM: Read Only Memory Y / C: Luminance and Chrominance

[0080]

According to the present invention, since the speed of the frame supplied to the decoder circuit is 1 × even during variable speed reproduction, the decoding speed of the decoder circuit is the same as that during normal reproduction, and The high-speed processing IC is not required as the decoder circuit, and the cost does not increase.

Further, when thinning out frames for variable speed reproduction, frames with a large volume level are left, so that it is possible to thin out important parts for understanding the contents in speech or movie dialogues. It is possible to suppress the decrease in understanding of the contents. Further, since the data can be thinned out by a method suitable for the contents to be reproduced at variable speed, it is possible to perform the variable reproduction at a wide range of contents which is easy to hear.

Further, even if the waveform is not continuous by thinning out frames, the volume at the discontinuous point is once reduced, so that noise caused by the discontinuity of the waveform does not stand out.

[Brief description of drawings]

FIG. 1 is a system diagram showing an embodiment of a recording / reproducing apparatus to which the present invention is applied.

FIG. 2 is a system diagram showing one embodiment of the present invention.

FIG. 3 is a diagram for explaining the operation of the circuit of FIG.

FIG. 4 is a diagram for explaining the operation of the circuit of FIG.

FIG. 5 is a system diagram showing another embodiment of the present invention.

FIG. 6 is a flowchart showing an example of a routine executed by the circuit of FIG.

FIG. 7 is a diagram for explaining the present invention.

FIG. 8 is a diagram for explaining the present invention.

[Explanation of symbols]

Reference numeral 11 ... Tuner, 12A to 13V ... Input terminal, 14 ... Input selector circuit, 15 ... A / D converter circuit, 16 ...
MPEG encoder circuit, 21 ... Y / C separation circuit, 22
... Input switching circuit, 23 ... NTSC decoder circuit, 2
4 ... Synchronous control circuit, 25 ... Pre-video signal processing circuit, 26
... MPEG encoder circuit, 31 ... multiplex /
Demultiplex circuit, 32 ... Buffer circuit, 33 ...
HDD, 41 ... MPEG-AV decoder circuit, 42 ... Switching circuit, 43 ... D / A converter circuit, 44 ... Output terminal, 51 ... Post video signal processing circuit, 52 ... OSD circuit, 53 ... NTSC encoder circuit, 54 and 55 …
Output terminal, 60 ... System controller, 411 ... Separation circuit, 412 ... Buffer circuit, 413 ... Decoder circuit body, 414 ... Parameter extraction circuit, 415 ... Scale factor determination circuit, 416 ... Rewrite control circuit

   ─────────────────────────────────────────────────── ─── Continued front page    F-term (reference) 5C052 AA01 AB02 AC01 CC01                 5C053 FA14 FA23 GB06 GB38 HA21                       JA01 KA24 KA25 KA26 LA07                 5D044 AB05 BC01 CC05 DE12 DE38                       DE49 FG07 FG23 GK12                 5J064 AA01 BB07 BC02 BC06 BC14                       BC18 BD03

Claims (8)

[Claims] 1. Digital audio data is made into one frame for each of a predetermined number of samples, data compression is performed for each frame, and information indicating the volume at the time of reproducing the digital audio data in each frame corresponds to the corresponding frame. For each frame, the frame is reproduced from the medium recorded integrally with the frame, and the reproduced frame is subjected to the data compression process complementary to the data compression to obtain the digital audio data. In the reproducing apparatus configured as described above, m / n times (m, n are integers,
/ N> 1) reproducing means, a first circuit for extracting the information indicating the sound volume for each frame from the frame reproduced by the reproducing means, and the first circuit for extracting the information. A second circuit for preferentially extracting n frames having digital audio data having a large volume from among m frames recorded continuously according to the information indicating the volume, and the data expansion process. And a decoder circuit for outputting the digital audio data to output the n frames fetched by the second circuit to the decoder circuit continuously for recording the digital audio data. / Reproduction device designed to obtain at n times speed. 2. The reproducing apparatus according to claim 1, wherein the data compression and the data expansion are in the MPEG system, and the information indicating the volume is a scale factor. 3. The reproducing apparatus according to claim 1, wherein the data compression and the data expansion are AC-3 (registered trademark) system, and the information indicating the volume is a coupling absolute exponent,
coupling exponents, channel exponents and low f
Replay device adapted to be one or more of requency effects channel exponents. 4. The reproducing apparatus according to claim 1, wherein the information indicating the volume extracted by the first circuit is information indicating the volume in a specific band, and a representative value of this information Of the above m frames,
N having the above-mentioned high-volume digital audio data
A playback device that preferentially takes out individual frames. 5. The reproducing apparatus according to claim 1, wherein when the n frames are continuously supplied to the decoder circuit, regarding the connection point of two frames which are not continuous at the time of recording, this connection point A reproduction device in which the information indicating the volume is rewritten so that the reproduced sound is faded out in the frame located before the frame and the reproduced sound is faded in in the frame located after the connection point. 6. Digital audio data is made into one frame for each of a predetermined number of samples, data compression is performed for each frame, and information indicating the volume at the time of reproducing the digital audio data in each frame corresponds to the corresponding frame. Each time, the frame is reproduced from the medium recorded integrally with the frame, and the reproduced frame is subjected to the data expansion processing complementary to the data compression to reproduce the digital audio data. In this case, the frame is recorded from the medium by m / n times (m, n
Is an integer and is reproduced at a speed of m / n> 1), information indicating the volume is extracted from the reproduced frame for each frame, and continuously recorded by the extracted information indicating the volume. Of the m frames, the n frames having the digital audio data with the high volume are preferentially extracted, and the n frames have the maximum processing speed of data expansion of m.
A reproducing method in which the digital audio data is continuously supplied to a decoder circuit which is not more than / n times to obtain the digital audio data at a recording speed of m / n. 7. The reproducing method according to claim 6, wherein the information indicating the extracted volume is information indicating the volume in a specific band, and the representative value of this information is used to represent the m frames. home,
N having the above-mentioned high-volume digital audio data
A playback method that preferentially takes out individual frames. 8. The reproducing method according to claim 6, wherein when the n frames are continuously supplied to the decoder circuit, the connection point of two frames which are not continuous at the time of recording is connected. A reproduction method in which the information indicating the above-mentioned volume is rewritten so that the reproduced sound is faded out in the frame located before and the reproduced sound is faded in in the frame located after the connection point.