JPH06324093A - Device for displaying spectrum of audio signal - Google Patents

Abstract

PURPOSE:To provide accurate resolution and to simplify the constitution of a device. CONSTITUTION:An encoded audio signal, which is inputted into an encoded- audio-signal input terminal, is sent into a time window means 12, and a plurality of time-division blocks are generated. The spectrums corresponding to a plurality of the time-division blocks are computed with a spectrum computing means 13. The computed spectrums are converted into the spectrum display signals with a display-signal forming means 14. Then, the spectrum display signals are displayed on a spectrum display means 15.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an audio signal spectrum display device for displaying an encoded audio signal for each spectrum.

[0002]

2. Description of the Related Art In order to perform recording without distortion on a magnetic tape, it is necessary to prevent the signal level from exceeding a predetermined threshold value. Therefore, in order to determine the true signal level at the time of recording, many commercially available cassette recorders and the like are equipped with a level meter. This level meter displays both left and right stereo channel signal levels. The level meter can also be thought of as a primitive spectrum display, displaying a time-varying signal level in a single frequency band over the entire frequency spectrum. Recently, this concept has been extended to display signal levels that change over time in several frequency bands. This spectrum display device is provided in a commercially available audio product or the like, and is displayed on a display device such as an LCD screen. This spectrum display device is also a general selling point for audio products regardless of whether or not it is used for recording.

In recent years, the audio and video industries are changing rapidly, and for example, the quality of the screen of a television is improved, and with the improvement of this screen, higher quality audio is required. Thus, the audio quality of television and video is beginning to approach that of commercial audio systems.
And by allowing the use of commercial audio systems to reorganize the soundtracks of video programs, for example, commercial audio,
Video and television devices should be more highly integrated. It is expected that more highly integrated systems will emerge in the future.

In the audio field, there has already been a rapid change, and so-called mini-discs (MD: MiniDiscs) and so-called digital compact cassettes (DDC: Digital Com).
Audio systems with reduced bit rates are being produced, such as the pact Cassette). These audio systems transmit audio information in the form of quantized coefficients that represent the energy of the voice within a given time and frequency domain. Current audio systems were originally conceived for audio applications, but in the future television and video audio may be transmitted using the above audio systems or systems similar to the above audio systems. Conceivable.

[0005]

In the above-mentioned spectrum display device, in order to display the coded data signal, the spectrum signal must be corrected to a spectrum signal required in each system, and the coded data signal must be corrected. What is needed is a device for converting a data signal into the frequency domain.

Therefore, in view of the above situation, the present invention provides a spectrum display device having a more accurate resolution than the conventional one and having a simple structure.

[0007]

SUMMARY OF THE INVENTION An audio signal spectrum display apparatus according to the present invention comprises time windowing means for selecting a time block of an encoded input audio signal,
Spectrum calculating means for calculating the display spectrum characteristic from the input signal of the time block of the windowed signal by the windowing means, and converting the spectrum characteristic calculated by the spectrum calculating means into a specific display signal The above-mentioned problem is solved by including display signal forming means for achieving the above.

Here, it is preferable to include a display device for displaying the display signal output by the display signal forming means. The display device may be an LCD display device or a television. Further, an output terminal for using a device connected to an external display device may be provided.

Further, it is conceivable that the encoded input audio signal is composed of a two-channel stereo audio signal.

Also, the encoded input audio signal comprises a coefficient corresponding to the energy of the signal in the time and frequency domain, the coefficient being divided uniformly or non-uniformly in time and frequency. -It may be a frequency coefficient. Further, the time block used by the time windowing means may correspond to the time division of the encoded input audio signal. Furthermore, the time block used by the time windowing means is characterized by a non-uniform time division of the encoded audio signal.

[0011] The above-mentioned encoded input audio signal is characterized by calculating a spectral characteristic or a plurality of characteristics by non-uniformly dividing time and frequency in order to non-uniformly dividing the time-frequency coefficient. To do.

The time-frequency coefficient of the audio signal is encoded by a single or a plurality of coefficients by a scale factor indicating the maximum value of the coefficient and a word length indicating the number of bits used to obtain each coefficient. It is characterized in that it is encoded by a block floating process.

Further, the time-frequency coefficient of the audio signal is generated by time-frequency conversion based on subband coding, and the coefficient of a continuous time block from a single subband is the single word length. And a block floating algorithm which is encoded by a scale factor.

The block floating algorithm is characterized in that it operates on a fixed time block so as to supply a scale factor to each subband at regular intervals.

Further, the block floating algorithm operates on the non-uniform time block so as to supply the value of the scale factor at a variable rate based on the spectral characteristic of the input audio signal. Characterize.

Further, the block floating algorithm is characterized in that it operates on a non-uniform time block so as to provide the values of the scale factor with different rates of different subbands.

The time-frequency coefficient of the audio signal is characterized by being generated by time-frequency conversion based on transform coding.

Also, the time-frequency coefficients of the audio signal may be generated by a time-frequency transform based on a combination of subband and transform coding such that the transform is adapted to each subband signal. Characterize.

Further, the time-frequency coefficients of the audio signal are encoded by a block floating algorithm which encodes blocks of adjacent frequency coefficients belonging to a single transform output with a single word length and scale factor. It is characterized by

Here, the block floating algorithm operates on a fixed time and frequency so as to provide a value of the scale factor corresponding to a fixed frequency range at fixed time intervals. Is characterized by.

Further, the block floating algorithm is characterized in that it operates on a variable time and frequency so as to supply a value of a scale factor according to a variable rate.

Further, in the block floating algorithm, the increased time resolution is reduced in frequency resolution, or the decreased time resolution is compensated by increased frequency resolution so that the scale factor value is supplied at a fixed rate. It is characterized in that it acts on a specific time and frequency.

The spectrum calculating means is characterized by extracting and using the scale factor from the encoded input audio signal to calculate the spectrum characteristic or a plurality of characteristics.

Further, the spectrum calculating means is characterized in that the spectrum characteristic or a plurality of characteristics is generated by directly outputting a scale factor corresponding to the calculated time block of each spectrum.

Here, the display device buffers the scale factor in order to produce a spectral output at regular intervals when the scale factor is supplied at a non-uniform rate along the frequency axis. It is characterized by including a time axis processing subsystem.

The display device also produces a new spectral output of each new scale factor or group of scale factors received, where the scale factor is provided by a non-uniform rate along the frequency axis. It is characterized by including a time axis processing subsystem.

[0027]

In the present invention, it is possible to easily display the spectrum signal having a more accurate resolution than the conventional one.

[0028]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT A preferred embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a view for explaining the basic concept of an audio signal spectrum display device according to the present invention, and FIG. 2 shows an example of using the audio signal spectrum display device according to the present invention. .

The encoded audio signal input terminal 11 of FIG.
The encoded audio signal is input from and is sent to the time windowing means 12. Since this encoded audio signal has a spectrum that changes with time, a plurality of time division blocks are generated in the time windowing means 12 from the encoded audio signal. The spectrum of each of the time division blocks is calculated by the spectrum calculating means 13 and sent to the display signal forming means 14. The display signal forming means 14 converts the calculated spectrum to form a spectrum display signal, and then the spectrum display means 15 displays the spectrum.

The spectrum display means 15 is an LCD display device or LE provided in a conventional audio device.
It is possible to use not only the D display device but also the television device 16 as shown in FIG. In this case, the audio signal from the encoded audio signal input terminal 11 is either an audio signal corresponding to a television signal or an audio signal received from an external audio device such as a radio or a compact disc player. is there.

Further, FIG. 3 shows an example of the basic structure of the spectrum display means 15 in the case where the spectrum display unit according to the present invention is physically separated from the processing block.

The connection section 17 for outputting the spectrum display signal in the audio signal conversion section 30 is connected to the television device 3.
1 for inputting the spectrum display signal in 1
The television device 31 includes a display unit 19 for displaying the spectrum display signal input via the connection unit 18 as a spectrum. The audio device 30 and the television device 31 are more highly integrated. Since the time windowing means 12, the spectrum calculating means 13, and the display signal forming means 14 in FIGS. 2 and 3 are the same as those in FIG. 1, the corresponding parts are designated by the same reference numerals and the description thereof will be omitted.

Next, a specific embodiment for calculating the spectrum generated by the time windowing means 12 and the spectrum calculating means 13 will be described later.

First, FIG. 4 shows a schematic configuration of the first embodiment of the present invention.

The coded audio signal input from the coded audio signal input terminal 41 generates a reduced bit rate as used in so-called MD (mini disc) and so-called DCC (digital compact cassette). It is an output signal from the audio signal encoding device.

Here, FIG. 5 shows a schematic configuration of encoding and decoding of a normal audio signal. The audio signal input from the audio signal input terminal 51 is encoded by the audio signal encoding means 52, transmitted to another audio device via the data transmission or recording means 53, or recorded in various recording media. To do. The data transmitted or recorded signal is decoded by the audio signal decoding means 54 and output from the audio signal output terminal 55.

FIG. 6 shows a schematic structure of the audio signal coding means 52 of FIG. Time-frequency analysis unit 61
Then, an audio signal in the time domain is divided into time-frequency components arranged in a space of time and frequency. This time-frequency component is sent to the bit allocation processing unit 62 and the spectrum quantization unit 63. The bit allocation processing unit 6
In No. 2, the quantization bit number is assigned based on the psychological acoustic model 64 so that the quantization noise generated by the quantization of the spectrum becomes minimum audible, and the quantization bit number is sent to the spectrum quantization unit 63. . In addition, each parameter such as a scale factor and word length is output. In the spectrum quantization unit 63, the time-frequency component output from the time-frequency analysis unit 61 is
After being quantized by the output from the bit allocation processing unit 62, a spectrum signal is output.

Usually, the spectrum quantizing unit 63 uses a quantizing method by block floating.
In this quantization method, first, time-frequency components are standardized by a scale factor and then quantized with a set word length. In this bit allocation process,
Determine the appropriate scale factor and word length for each time block. This scale factor is generally a value selected from among a plurality of fixed parameters. Also, the scale factor selected for each block is the minimum scale factor, which is greater than the magnitude of all time and frequency coefficients in the block.

Thus, this scale factor is a value that approximates the energy of the audio signal within the time and frequency range of the block of time and frequency coefficients. Therefore,
Multiple scale factors corresponding to the entire frequency range at a particular point in time can be used as a good approximation of the spectral signal.

Actually, the operation by the audio signal coding means 52 is performed by the time windowing means 12 and the spectrum calculating means 13 of FIG. In FIG. 4, each parameter by the block floating algorithm is extracted by the encoded data extracting means 42, and the data of this scale factor is sent to the spectrum display means 15 through the display signal forming means 14. The encoded data extraction means 42 is a part of the audio signal decoding means 54 of FIG. In the method according to the first embodiment, the spectral signal which changes with time is highly efficiently and easily calculated and displayed.

Next, FIG. 7 shows a schematic configuration of the second embodiment of the present invention.

In the spectrum display apparatus according to the second embodiment, a coded audio signal having a variable time resolution is input to the coded audio signal input terminal 41 (first case), or a display signal is formed. Means 14
And the calculation of the spectrum signal when the time resolution of the spectrum display means 15 is different from the resolution of the input encoded audio signal (second case) will be considered.

According to the second embodiment, the coded audio signal input to the coded audio signal input terminal 41 has the scale factor, word length, coefficient, etc. extracted by the coded data extracting means 42. It is sent to the time axis processing means 43. By holding the respective values in the time axis processing means 43, the scale factor data at constant time intervals can be converted into the display signal forming means 1.
Supply to 4. The display signal forming means 14 forms a spectrum display signal, and the spectrum display signal is supplied to the spectrum display means 15 to display a spectrum.

The first case is the case of an input audio signal having a variable time resolution, and occurs when the block distributed by the time-frequency analysis unit 61 in the audio signal encoding means 52 is variable. . It also occurs when the time-frequency analysis unit 61 has an analysis window that changes with time. This method is used to prevent pre-echoes that occur in the amplitude part of the signal due to the analysis window being too short during the amplitude of the signal with a large amplitude change.

In the first case, a variable rate scale factor is input to the time axis processing means 43 of FIG. 7, converted to a fixed rate required by the display signal forming means 14 and output. . Here, if several scale factors are received in a single frequency band, the time axis processing means 43 selects the most approximate scale factor such as an average value or a maximum value. Also, if a given frequency band does not receive a new scale factor within the current time block, the previous scale factor will be used.

Furthermore, in the first case, if the display rate is greater than the maximum rate at which a new scale factor was received, the time base processing means 43
The spectrum display means 15 acts as a buffer to ensure that it corresponds to each new scale factor. Thus, the spectrum display means 15 quickly responds to the rate of increased scale factor produced when the frequency band is input in the short analysis mode. The above-mentioned operation generally occurs in a portion where the amplitude of the signal changes drastically, and the spectrum display means 15 automatically adapts to a sudden increase in energy in the portion where the amplitude changes drastically. This adapted result is a desired feature that is not valid for fixed window spectral displays.

The second case is the case of an input audio signal having a time resolution different from that of the spectrum display means 15, and the time axis processing means 43 receives all the received signals during each display cycle. Holds the scale factor. If no new scale factor is received, i.e. the display cycle is shorter than the signal analysis window, the time base processing means 43 outputs one of the scale factors of the previous display cycle. Further, when more scale factors than the one scale factor are received, the time axis processing means 43.
Selects the scale factor that is the closest to, ie, maximum or equivalent to, each frequency band and outputs this scale factor.

As mentioned above, constant spectral coefficients are transmitted and displayed by the spectral display device of the present invention.

Furthermore, in order to obtain the spectrum used in the present invention, a high efficiency compression coding method can be used as the third embodiment. Therefore, a specific example of this high-efficiency compression encoding will be described in detail. That is, an input digital signal such as an audio PCM signal is subjected to band division coding (SBC),
A technique for highly efficient coding using each technique of adaptive transform coding (ATC) and adaptive bit allocation will be described with reference to FIG. 8 and subsequent figures.

In the concrete high-efficiency coding apparatus shown in FIG. 8, the input digital signal is divided into a plurality of frequency bands, and the bandwidths of the two adjacent lowest bands are the same, and are higher in the higher frequency bands. The wider the frequency band, the wider the bandwidth is selected, and the spectrum data on the frequency axis that is obtained by performing orthogonal transformation for each frequency band is used in the low frequency range. ), In the high and middle frequency band, the critical bandwidth is subdivided in consideration of the block floating efficiency, and each bit is adaptively bit-assigned and coded. Usually, this block is the quantization noise generation block. Furthermore, in the embodiment of the present invention, the block size (block length) is adaptively changed according to the input signal before the orthogonal transformation, and the floating process is performed for each block.

That is, in FIG. 8, for example, when the sampling frequency is 44.1 kHz, 0 to 0 is input to the input terminal 100.
A 22 kHz audio PCM signal is supplied.
This input signal is divided into 0 to 11 kHz band and 11 by a band division filter 101 such as a so-called QMF filter.
It is divided into the band of kHz to 22 kHz (high range), and it is 0-1
A signal in the 1 kHz band is similarly divided into a 0 to 5.5 kHz band (low band) and a 5.5 kHz to 11 kHz band (middle band) by a band dividing filter 102 such as a so-called QMF filter. The signals of each band from the band division filters 101 and 102 are sent to the orthogonal transform block size determination circuit 106, and the block size is determined for each band.

This orthogonal transform block size determination circuit 10
6, the block size length is 11.6 m, for example.
Based on the length of s, this length becomes the maximum block size. If the signal is quasi-stationary in time, the maximum block size is 11.6 which is the orthogonal transform block size.
When the frequency resolution is increased by selecting ms and the signal is non-stationary in time, the orthogonal transform block size is further divided into four in the band of 11 kHz or less, and
In the band of 1 kHz or higher, the time resolution is improved by dividing the orthogonal transform block size into eight.

As a method of dividing the above-mentioned input digital signal into a plurality of frequency bands, there is, for example, a QMF filter, and 1976 RECrochiere Digital Coding
fSpeech in Subbands Bell Syst.Tech. J. Vol.55, No.8
1976. Also, ICASSP 83, Boston Poly
phase Quadrature Filters-A New Subband CodingTechn
ique Joseph H. Rothweiler describes a technique for partitioning filters with equal bandwidth.

In FIG. 8, the band division filter 101,
The output from 102 is supplied to each orthogonal transform circuit 103, 104, 105 for each signal in each band. At the same time, the block size determined by the orthogonal transform block size determination circuit 106 is determined by the orthogonal transform circuit 10
3, 104, 105, and the outputs from the band division filters 101, 102 are divided into blocks according to the block size and subjected to orthogonal transform processing. FIG. 9 shows the orthogonal transform block size, which is 11.6 ms (long mode) or 2.9 ms in the low and middle frequencies.
Select either (Short mode), 1 in high range
Select either 1.6 ms (long mode) or 1.45 ms (short mode). The determined orthogonal transform block size information is taken out from the terminal 111 of FIG. 8 and sent to the decoding circuit.

Here, as the above-mentioned orthogonal transform, for example, the input audio signal is divided into blocks in a predetermined unit time (frame), and fast Fourier transform (FF) is performed for each block.
T), Discrete Cosine Transform (DCT), Modified Discrete DCT Transform (MDCT), etc. are used to transform the time axis into the frequency axis. ICASSP for MDCT
1987 Subband / Transform Coding Using Filter Bank D
esigns Based on TimeDomain Aliasing Cancellation
JPPrincen ABBradley Univ. Of SurreyRoyal Me
lbourne Inst. of Tech.

Next, FIG. 10 shows a block circuit showing a schematic configuration of a specific example of the bit distribution calculating circuit 107.
In FIG. 10, input terminal 121 is supplied with spectrum data on the frequency axis from each of the orthogonal transform circuits 103, 104 and 105.

The input data on the frequency axis is sent to the energy calculation circuit 122 for each band, and the energy of each divided band considering the masking amount, the critical band and the block floating is, for example, each energy in the band. It can be obtained by calculating the sum of amplitude values.
Instead of the energy for each band, a peak value, an average value, etc. of the amplitude value may be used. As an output from the energy calculation circuit 122, for example, the spectrum of the sum value of each band is shown as the spectrum SB in FIG. However, in FIG. 11, in order to simplify the illustration, the number of division bands in consideration of the masking amount, the critical band, and the block floating is represented by 12 bands (B1 to B12).

Here, in order to consider the influence of so-called masking on the spectrum SB, a convolution process is performed such that the spectrum SB is multiplied by a predetermined weighting function and added. Therefore, the output of the energy calculation circuit 22 for each band, that is, each value of the spectrum SB is sent to the convolution filter circuit 123. The convolution filter circuit 123 includes, for example, a plurality of delay elements that sequentially delay input data and a plurality of multipliers that multiply outputs from these delay elements by a filter coefficient (weighting function) (for example, 25 corresponding to each band). Number of multipliers), and a sum adder that sums the outputs of the multipliers. By this convolution processing, FIG.
The sum of the parts shown by the dotted line in the figure is taken.

The masking means a phenomenon in which one signal is masked by another signal and becomes inaudible due to human auditory characteristics. The masking effect depends on the audio signal on the time axis. There are a time axis masking effect and a simultaneous time masking effect by a signal on the frequency axis. Due to these masking effects, even if there is noise in the masked portion, this noise cannot be heard. Therefore, in the actual audio signal, the noise within the masked range is regarded as an acceptable noise.

Here, a specific example of the multiplication coefficient (filter coefficient) of each multiplier of the convolution filter circuit 123 will be described. When the coefficient of the multiplier M corresponding to an arbitrary band is 1, the multiplier is M-1 gives a coefficient of 0.15, multiplier M-2 gives a coefficient of 0.0019, and multiplier M-3 gives a coefficient of 0.0000.
086, multiplier M + 1 gives a coefficient of 0.4, multiplier M + 2
By multiplying the output of each delay element by a coefficient of 0.06 with a coefficient of 0.007 with a multiplier M + 3, the convolution processing of the spectrum SB is performed. However, M is an arbitrary integer of 1 to 25.

The output from the convolution filter circuit 123 is sent to the subtractor 124. The subtractor 124 calculates a level α corresponding to an allowable noise level described later in the convoluted area. The level α corresponding to the permissible noise level (permissible noise level) is a level at which the critical noise band becomes the permissible noise level for each band by performing inverse convolution processing, as described later. is there. here,
The subtractor 124 is supplied with an allowance function (function expressing a masking level) for obtaining the level α. The level α is controlled by increasing or decreasing this allowance function. The permissible function is supplied from the (n-ai) function generating circuit 125 described below.

That is, the level α corresponding to the allowable noise level can be obtained by the following equation (1), where i is the number given in order from the low band of the critical band. α = S- (n-ai) (1) In this equation (1), n and a are constants, a> 0, S is the intensity of the convolution-processed Bark spectrum, and (1)
In the formula, (n-ai) is the allowable function. In this embodiment, n
= 38, a = 1, there is no sound quality deterioration at this time,
Good encoding can be performed.

In this way, the level α is obtained, and this data is transmitted to the divider 126. The divider 126 is for deconvoluting the level α in the convolved region. Therefore,
By performing this inverse convolution processing, the masking spectrum can be obtained from the level α. That is, this masking spectrum becomes the allowable noise spectrum. Although the above-mentioned inverse convolution processing requires a complicated operation, in the present embodiment, the inverse convolution is performed using the simplified divider 126.

Next, the masking spectrum is transmitted to the subtractor 128 via the synthesizing circuit 127. Here, the output from the energy detection circuit 122 for each band, that is, the spectrum SB described above, is input to the subtractor 128.
Are supplied via the delay circuit 129. Therefore,
By performing the subtraction operation of the masking spectrum and the spectrum SB in the subtractor 128, as shown in FIG. 12, the spectrum SB is masked below the level indicated by the level of the masking spectrum MS.

The output from the subtracter 128 is taken out through the output terminal 131 through the allowable noise correction circuit 130, and, for example, RO in which the allocated bit number information is stored in advance.
M, etc. (not shown). This ROM or the like is assigned to each band according to the output (the level of the difference between the energy of each band and the output of the noise level setting means) obtained from the subtraction circuit 128 via the allowable noise correction circuit 130. Outputs bit number information. This allocation bit number information is sent to the adaptive bit allocation encoding circuit 108 in FIG. 8 so that each spectrum data on the frequency axis from each orthogonal transformation circuit 103, 104, 105 is allocated to each band. It is quantized by a number.

That is, in summary, in the adaptive bit allocation encoding circuit 108 of FIG. 8, the difference between the masking amount, the energy of each divided band considering the critical band and the block floating, and the output of the noise level setting means is calculated. The spectrum data for each band is quantized by the number of bits assigned according to the level.
The delay circuit 129 is provided in order to delay the spectrum SB from the energy detection circuit 122 in consideration of the delay amount in each circuit before the synthesis circuit 127.

By the way, at the time of synthesizing by the synthesizing circuit 127 described above, data indicating a so-called minimum audible curve RC which is the human auditory characteristic as shown in FIG. The masking spectrum MS can be combined. In this minimum audible curve, if the absolute noise level is below this minimum audible curve, the noise will not be heard. Even if the coding is the same, the minimum audible curve differs depending on, for example, the difference in reproduction volume at the time of reproduction, but in a realistic digital system, for example, how to enter music into a 16-bit dynamic range is very different. Therefore, if the quantization noise in the most audible frequency band around 4 kHz is not heard, it is considered that the quantization noise below the level of the minimum audible curve is not heard in other frequency bands.

Accordingly, assuming that the system is used in such a manner that noise near the 4 kHz of the word length possessed by the system cannot be heard, the minimum audible curve RC and the masking spectrum MS are combined together to obtain an allowable noise level. In this case, the allowable noise level in this case can be up to the shaded portion in FIG. In this embodiment, the level of 4 kHz of the minimum audible curve is set to the minimum level equivalent to 20 bits, for example. Further, FIG. 13 also shows the signal spectrum SS at the same time.

Further, in the allowable noise correction circuit 130,
The allowable noise level in the output from the subtractor 128 is corrected based on, for example, the equal loudness curve information sent from the correction information output circuit 133. Here, the equal loudness curve is a characteristic curve relating to human auditory characteristics, and is obtained by, for example, obtaining the sound pressure of sound at each frequency heard at the same loudness as a pure tone of 1 kHz and connecting them with a curve. Also called sensitivity curve. Further, this equal loudness curve is the minimum audible curve R shown in FIG.
It draws a curve substantially the same as C. In this equal loudness curve, for example, in the vicinity of 4 kHz, even if the sound pressure is reduced by 8 to 10 dB from 1 kHz, it sounds as loud as 1 kHz, and conversely, in the vicinity of 50 Hz, it is not higher than the sound pressure at 1 kHz by about 15 dB. It doesn't sound the same. Therefore, it is understood that it is preferable that the noise (allowable noise level) exceeding the level of the minimum audible curve has a frequency characteristic given by a curve corresponding to the equal loudness curve. From this, it can be seen that correcting the permissible noise level in consideration of the equal loudness curve is suitable for human hearing characteristics.

Here, as the correction information output circuit 133, the detection output of the output information amount (data amount) at the time of quantization in the adaptive bit allocation encoding circuit 108 of FIG. 8 and the bit of the final encoded data The allowable noise level may be corrected based on the information on the error from the rate target value. This is because the total number of bits obtained by performing temporary adaptive bit allocation in advance for all bit allocation unit blocks is a fixed number of bits (target value) determined by the bit rate of the final encoded output data. May have an error with respect to, and bit allocation is performed again so that the error may be zero. That is, when the total allocated bit number is smaller than the target value, the difference bit number is allocated to each unit block and added, and when the total allocated bit number is larger than the target value, the difference bit number is set in each unit. Allocate to blocks and delete.

In order to do this, an error of the total allocated bit number from the target value is detected, and the correction information output circuit 133 corrects the allocated bit number according to the error data. Output the data. Here, when the error data indicates a bit number shortage, it can be considered that the data amount is larger than the target value because a large number of bits are used per unit block. Further, when the error data is data indicating a surplus of the number of bits, it can be considered that the number of bits per unit block is small and the amount of data is smaller than the target value. Therefore, the correction information output circuit 133 outputs the correction value for correcting the allowable noise level in the output from the subtractor 128 based on the error data, for example, based on the information data of the equal loudness curve. Data will be output. By transmitting the correction value as described above to the allowable noise correction circuit 130, the subtracter 128
The allowable noise level from is corrected. In the system as described above, the data obtained by processing the orthogonal transform output spectrum by the sub information as the main information,
As the sub information, a scale factor indicating a block floating state and a word length indicating a word length are obtained and sent from the encoder to the decoder.

An effective bit allocation method different from the above-mentioned bit allocation method will be described below.

The operation of the adaptive bit allocation circuit will be described with reference to FIG. Orthogonal transform output, for example MDCT output,
This MDCT output has the energy for each divided band in the energy calculation circuit 301 for each band, that is, a band obtained by further dividing the critical band into a plurality of bands in the critical band or the high band, that is, a so-called block floating band. It is calculated. A peak value, an average value or the like of the amplitude may be used instead of the energy of each band.

By the way, assuming that the total number of bits 305 that can be used for transmission or recording by expressing the MDCT coefficient which is an orthogonal transform output is 1 kbit / block, a fixed bit allocation pattern 307 using this 1k bit is obtained. Form. A plurality of bit allocation patterns for fixed bit allocation are prepared, and various selections can be made depending on the nature of the signal. This bit allocation pattern has 1
There are various patterns in which the bit amount of a short time block corresponding to k bits is distributed to each frequency. In particular, in the present embodiment, a plurality of patterns having different bit allocation ratios in the low and middle frequencies are prepared. Here, as the signal size is smaller, a pattern with a smaller amount of allocation to the high frequency band is selected. In this way, as the small signal size, the signal size of the entire band can be used,
Furthermore, the output of the non-blocking frequency division circuit using a filter or the like, or the orthogonal transform output, for example, the MDCT output is used. An energy-dependent bit allocation pattern 306 is determined from the energy of each band. This energy-dependent bit allocation pattern 306 is
For example, the larger the energy of the band is, the more bits are allocated and allocated.

The division ratio between the bit allocation of the fixed bit allocation pattern 307 and the spectrum-dependent bit allocation of each band is determined by an index (tonality) indicating the smoothness of the signal spectrum. In the present embodiment, the spectrum smoothness calculation circuit 302 calculates a value obtained by dividing the sum of absolute values of differences between adjacent values of the signal spectrum by the sum of the signal spectrum, and uses this value as an index (tonality). ing. When the tonality is determined, the bit division rate determination circuit 304 determines the division rate. This division ratio is a value for changing the weighting of fixed bit allocation and energy-dependent bit allocation.

The value of the fixed bit allocation is the multiplier 309.
Then, the value of the bit allocation depending on the energy of each band (critical band or a band in which the critical band is further subdivided in the high band) is multiplied by the above allocation ratio in the multiplier 308,
These two values are added in the adder circuit 310, taken out from the terminal 311, and used in the quantization and encoding.

The state of bit allocation at this time is shown in (a) of FIG. 15 and (a) of FIG. 16, and the state of quantization noise corresponding to this is shown in (b) of FIG. 15 and (b) of FIG. Shown in.

In FIGS. 15A and 16A, the noise level due to the fixed bit allocation in the signal level a is shown by b, and the noise reduction due to the signal level dependent is shown by c. FIG. 15 shows the case where the spectrum of the signal is relatively flat, and bit allocation with a large amount of fixed bit allocation helps to have a large signal-to-noise ratio in the entire band. However, relatively low bit allocations are used in the low and high frequencies. This is because the importance of the low range and the high range is auditorily small. At the same time, the noise level in the band in which the signal size is large is selectively lowered by the bit amount corresponding to the signal level shown by the diagonal lines in FIG. 15B for the bit allocation depending on the signal level. However, if the signal spectrum is relatively flat, this selectivity will also work over a relatively wide band.

On the other hand, as shown in FIG. 16, when the signal spectrum shows a high tonality, the signal level indicated by the diagonal lines in FIG. 16B for a large amount of signal level-dependent bit allocation. Due to the amount of dependent bits, the quantization noise reduction is used to reduce noise in a very narrow band. This achieves improved performance with isolated spectrum input signals. At the same time, the noise level in a wide band is non-selectively lowered by the bit allocation by a slight fixed bit allocation.

Again, referring to FIG. 8, the adaptive bit allocation encoding circuit 108 will be described. In this embodiment, for example, there are two kinds of bit rate modes, for example, A mode is 128 kbps / channel and B mode is A.
Half the mode, 64 kbps / channel. In addition, the present embodiment is not limited to the two types of modes, and it is possible to have a plurality of modes.

First, the encoding method in the A mode will be described. 17 and 18 show a specific example of block floating band division in the A mode. FIG. 17 shows the case where the orthogonal transform block size is 11.6 ms, and FIG. 18 shows the case where the orthogonal transform block size is divided into 4 in the low and middle regions and 8 in the high region. The number of block floating bands is the same and is divided into 52 bands. Further, looking at each band output from the band division filter, there are 20 block floating in the low band and 16 block floating in the middle and high band, respectively. Since this number is determined regardless of the orthogonal transform block size, the orthogonal transform is performed. There is no problem if the conversion block size changes independently for each band.

For example, in the case where the low-frequency band has a block size obtained by dividing 11.6 ms into four blocks and the mid-high band has a block size of 11.6 ms, the block floating band is
If the low band is divided as shown in FIG. 18 and the middle and high bands are divided as shown in FIG. 17, the total number of bands is 52. The adaptive bit allocation coding circuit 108 is provided with scale factor and word length information for each of the 52 block floating bands, and the spectrum data is quantized in accordance with the given scale factor and word length. And encoded. The encoded data is taken out from the terminal 110 and recorded or transmitted.

Next, the B-mode coding method will be described. Since the bit rate in B mode is half that in A mode, the amount of sub information (scale factor, word length, etc.) does not change when encoded in the same manner as in A mode, and the main information (spectral data) As a result, the amount of sub information in the total amount of information increases, and the coding efficiency decreases, as compared with the A mode. When the bit rate is halved, it is desirable to reduce not only the main information amount but also the sub information amount by half or less. In the present embodiment, in order to reduce the amount of sub information in B mode to half that in A mode, the value of sub information is shared between two blocks that are temporally adjacent to each other, thereby reducing the amount of sub information. Has achieved. That is, since the amount of sub information in A mode is basically equal to the number of block floating bands, 52 /
Although it is 11.6 ms, in the B mode, since the time axis direction of the block floating band is expanded, it becomes 52 pieces / 23.2 ms, and comparing the sub information amounts in the same time, it is compared with the A mode. It is half the amount.

FIGS. 19, 20, and 21 show specific examples of block floating band division in the B mode.

FIG. 19 shows a case where the orthogonal transform block sizes of two blocks temporally adjacent to each other are both in the long mode, and the region surrounded by the solid line is the orthogonal transform block and the region shown by diagonal lines. Indicates one block floating band. That is, this block floating band has two bands adjacent to each other in the time axis direction of the A mode Flock floating band in FIG. 17, and the band division in the frequency axis direction is exactly the same as that in FIG.

FIG. 20 shows the case where the orthogonal transform block sizes of two blocks that are temporally adjacent are both in the short mode. As in FIG. 19, the region surrounded by the solid line is the orthogonal transform block and the diagonal line. The area indicated by represents one block floating band. That is, this block floating band is a combination of two adjacent bands in the time axis direction of the A mode block floating band in FIG. 18, and the band division in the frequency axis direction is exactly the same as in FIG.

FIG. 21 shows a case where two blocks that are temporally adjacent have different orthogonal transform block sizes, that is, a combination of a short mode and a long mode.
Similarly, a region surrounded by a solid line represents an orthogonal transform block, and a region shaded by a diagonal line represents one block floating band. A block in which the orthogonal transform block size is the short mode (0-11.
For the 6 ms middle band and the low and high bands of 11.6 to 23.2 ms), the orthogonal transform block sizes of the two blocks as shown in FIG. 20 are the same as those in the short mode. That is, two adjacent bands in the time axis direction of the block floating band of the A mode in FIG. 18 are combined into one, and the band division in the frequency axis direction is exactly the same as in FIG. On the contrary, a block whose orthogonal transform block size is the long mode (0 in FIG. 21).
~ 11.6ms low and high range and 11.6 ~ 23.2m
In the middle range of s), the bands cannot be put together in the time axis direction, so that the two adjacent bands in the frequency axis direction are exceptional.
The bands are combined into one, and the band division in the time axis direction is exactly the same as that in FIG.

As described above, in order to reduce the number of sub-information in the B mode to half that in the A mode, the adjacent block floating bands in the time axis direction or the frequency axis direction are made common, resulting in bit The coding efficiency is improved by preventing the extreme decrease of the main information due to the decrease of the rate.

FIG. 22 shows a specific example of the adaptive bit allocation encoding circuit in the B mode, and the terminal 40
Orthogonal transform block size information is given to 1, and spectrum data (MDCT coefficients) for the two blocks is given to the terminal 402.

Scale factor A4 set for each band by block floating band division for A mode
03 in the scale factor reset circuit 405,
As described above, the values of the two floc floating bands to be made common are put together, and the scale factor B for B mode is reset. Usually, a larger scale factor A is selected from the two scale factors A,
A common scale factor A is used. Similarly, the word length A 404 set for each band in the block floating band division for A mode is reset to the word length B for B mode in the word length reset circuit 406. When sharing word lengths,
For example, of the two word lengths A, the larger word length A is selected. Furthermore, the average value of the two word lengths A or the like may be used. The scale factor A and the word length A are each 2 blocks (2
Information of 3.2 ms) is sent as one unit to the scale factor resetting circuit 405 and the word length resetting circuit 406. The word length reset by the word length reset circuit 406 is corrected by the total bit number correction circuit 407 for the error of the total bit number caused by the reset. The scale factor B and the word length B which are reset are both the quantizer 4
08 and the encoder 409, and is used when the spectrum data is quantized. The spectrum data that has been quantized by the quantizer 408 and the encoder 409 and coded is taken out from the terminal 410 as coded data B.

The above embodiment is an example of the present invention.
It goes without saying that various other configurations can be adopted without departing from the scope of the present invention.

[0093]

As is apparent from the above description, in the audio signal spectrum display apparatus according to the present invention, the time windowing means for selecting the time block of the encoded input audio signal and the windowing means. Spectrum calculation means for calculating spectrum characteristics for display from the input signal of the time block of the signal windowed by, and display for converting and forming the spectrum characteristics calculated by the spectrum calculation means into a specific display signal. By including the signal forming means, it is possible to display a spectrum signal with higher precision and higher resolution than the conventional spectrum signal with a simpler configuration.

[Brief description of drawings]

FIG. 1 is a diagram for explaining a basic concept of a spectrum display device for audio signals according to the present invention.

FIG. 2 is a diagram showing an example of using the spectrum display device for audio signals according to the present invention.

FIG. 3 is a diagram showing an application example using a spectrum display device for audio signals according to the present invention.

FIG. 4 is a diagram showing a schematic configuration of a first embodiment of the present invention.

FIG. 5 is a diagram showing encoding and decoding of a general audio signal.

FIG. 6 is a diagram showing a schematic configuration of an audio signal encoding means according to the present invention.

FIG. 7 is a diagram showing a schematic configuration of a second embodiment of the present invention.

FIG. 8 is a block circuit diagram showing a specific example of a high-efficiency compression encoding circuit.

FIG. 9 is a diagram showing the structure of an orthogonal transform block at the time of bit compression.

FIG. 10 is a block circuit diagram showing a specific example of a bit allocation calculation circuit.

FIG. 11 is a diagram showing spectra of bands divided in consideration of each critical band and block floating.

FIG. 12 is a diagram showing a masking spectrum.

FIG. 13 is a diagram in which a minimum audible curve and a masking spectrum are combined.

FIG. 14 is a block circuit diagram showing a specific example of an adaptive bit allocation circuit.

FIG. 15 is a diagram showing a signal level and quantization noise state in which bits are assigned by an adaptive bit assignment circuit.

FIG. 16 is a diagram showing a second state of the signal level and quantization noise bit-allocated by the adaptive bit allocation circuit.

FIG. 17 is a diagram specifically showing a frequency and time by block floating band division in A mode.

FIG. 18 is a second diagram specifically showing the frequency and time by block floating band division in A mode.

FIG. 19 is a diagram specifically showing frequency and time by block floating band division in B mode.

FIG. 20 is a second diagram specifically showing the frequency and time by block floating band division in B mode.

FIG. 21 is a third diagram specifically showing frequency and time by block floating band division in B mode.

FIG. 22 is a block circuit diagram showing a specific example of a B-mode adaptive bit allocation encoding circuit.

[Explanation of symbols]

11: coded audio signal input terminal 12: time windowing means 13: spectrum calculation means 14: ........ Display signal forming means 15 ........ Spectrum display means 16 .................. TV device 17, 18, .. 19 Display unit 30 Audio device 31 Television device 42 Encoded data extraction means 43 ... Time axis processing means 52 ... Audio signal coding means 54 ... Audio signal decoding Means of conversion

Claims (10)

[Claims] 1. A spectrum display device for displaying a frequency spectrum of an audio signal, comprising: time windowing means for selecting a time block of an encoded input audio signal; and time of a signal windowed by the windowing means. The block includes spectrum calculating means for calculating a display spectrum characteristic from the input signal of the block, and display signal forming means for converting and forming the spectrum characteristic calculated by the spectrum calculating means into a specific display signal. A spectrum display device for audio signals. 2. A spectrum display device for an audio signal according to claim 1, further comprising a display device for displaying a display signal output by said display signal forming means. 3. A spectrum display device for audio signals according to claim 2, wherein said display device comprises an LCD display device. 4. A spectrum display device for audio signals according to claim 2, wherein said display device comprises a television device. 5. An output terminal for using a device connected to an external display device is provided.
A spectrum display device for an audio signal according to claim 1. 6. A spectrum display device for audio signals according to claim 1, wherein the encoded input audio signal comprises a two-channel stereo audio signal. 7. The coded input audio signal comprises coefficients corresponding to the energy of the signal in the time and frequency domain.
7. A spectrum display device for audio signals according to 4, 5, or 6. 8. A spectrum display device for audio signals according to claim 7, wherein said coefficient is a time-frequency coefficient uniformly divided by time and frequency. 9. The apparatus of claim 7, wherein the coefficient is a time-frequency coefficient which is non-uniformly divided by time and frequency. 10. A spectrum display device for audio signals according to claim 1, wherein the time block used by the time windowing means is a time division of the encoded input audio signal.