JP3152114B2 - Audio signal encoding device and decoding device - Google Patents

Abstract

PURPOSE: To exclusive feature points not so important in terms of audition from a transmission object and to perform highly efficient encoding by extracting the feature points by using the correlation function between the sub band signal and the convex part of a trigonometric function. CONSTITUTION: Respective feature point extraction circuits 2-1-2-32 are provided with cosine functions b1(t)=cos(θt+τ) and b2(t)=-cos(θt+τ) defined as the convex function of a positive side and a negative side respectively. In the feature point extraction circuit 2-n, the correlation values between the sub band signals S1-S32 and the respective cosine functions b1(t) and b2(t) are obtained. In this case, the respective correlation functions for which a time different τ is a variable are changed approximately in a triangular wave shape corresponding to the amplitude of the sub band signal Sn. Then, the feature point extraction circuit 2-n sets the maximum value or minimum value of the sub band signal Sn in a time band where the correlation value exceeds a prescribed threshold value as the feature point.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an audio signal encoding apparatus and an audio signal decoding apparatus, and enables a large amount of data compression and decoding of encoded data by high-efficiency encoding considering characteristics of an audio signal. The present invention relates to an improvement for improving transmission efficiency and reducing the capacity of a storage medium.

[0002]

2. Description of the Related Art Conventionally, in encoding an audio signal, predictive encoding or ADPCM (Adaptive Modulation) has been used as a typical method.
ive Differential Pulse Code Modulation) and the like, and transmission efficiency is improved by data compression. Here, the predictive coding method is a method of predicting and estimating the current sample value from some past sample values, encoding the difference between the true sample value and the predicted value (prediction error), and transmitting the encoded value. For a normal audio signal, a simple difference PCM using a value one sample before as a predicted value is 8 to 9 per sample.
Compression on the order of bits is possible. The ADPCM system is a system in which an adaptive step width is introduced into the above-described differential PCM and the quantization step width is changed in accordance with the properties of signals in the vicinity. Widely used. However, for a signal having a large time-series change such as a tone signal, the prediction error becomes large, and a sufficient compression ratio cannot be obtained by the above method in many cases.

When audio signals are sampled, the perceptual importance of individual quantized data is not uniform. That is, the extreme value of the time-series amplitude change has a greater importance, and the importance decreases when the amplitude is constantly increasing or decreasing. Therefore, the encoding and
Although the transmission method is applied to decoding, the encoding device considers the maximum value and the minimum value of the change in the input signal as feature points, and uses only signal level information and time information of the feature points as feature point information. It is conceivable to apply a method (disclosed in JP-A-6-189288 etc.) that restores the original signal by interpolating between each feature point with an appropriate function on the decoding device side. . Specifically, on the encoding device side, the input signal is divided into a plurality of sub-band signals, the feature points are extracted for each sub-band signal, and feature point information for each sub-band signal is multiplexed. The decoding device side separates the transmission signal for each sub-band, and performs time-series rearrangement based on feature point information and performs interpolation processing using an appropriate function to restore each sub-band signal. Then, a decoded signal is obtained by synthesizing each sub-band signal.

[0004]

The above-mentioned feature point information encoding / transmission system can eliminate redundancy due to data of small importance between extreme values without substantially impairing the auditory characteristics of the audio signal. It seems that significant compression of the transmitted data is possible. However, this method has not been adopted for encoding / decoding audio signals. The reason is that in the case of image signals, the number of feature points is reduced due to strong correlation not only between frames and fields but also before and after signals on a scanning line. This is because the correlation is weak and the number of feature points is extremely large, so that highly efficient coding cannot be realized. That is, when the sub-band signal as shown in FIG.
It is extracted as n ~ Pn + 10, but is aurally Pn, Pn + 3,
Only Pn + 7, Pn + 8, Pn + 11 have significance, the other Pn + 1, Pn
n + 2, Pn + 4, Pn + 5, Pn + 6, and Pn + 10 are of relatively low importance, and the latter feature point information greatly increases the amount of transmission data.

On the other hand, the method of extracting the maximum value and the minimum value of a signal as feature points can be said to be a method suitable for the characteristics of an audio signal as described above. The reason why a sufficient data compression ratio cannot be obtained by this method is that information on feature points that are not perceptually important is to be transmitted. Therefore, the present invention devises the extraction of the feature points of each subband signal, eliminates the feature points that are not so important in hearing from the transmission target, realizes highly efficient encoding, and furthermore, based on the feature point information. It has been created for the purpose of providing an audio signal encoding device and a decoding device capable of rationally decoding an original signal.

[0006]

According to a first aspect of the present invention, there is provided a band dividing unit for dividing an input audio signal into a plurality of sub-band signals, and a plurality of sub-band signals are provided corresponding to respective divided outputs of the band dividing unit. A feature point extracting means for extracting a feature point from the signal waveform of the sub-band signal to obtain signal level information and time information of each feature point; and An audio signal encoding apparatus comprising: a quantizing means for quantizing the feature point information obtained by the point extracting means; and a transmitting means for multiplexing and transmitting the quantized data from the quantizing means. Each feature point extracting means calculates a correlation function between the convex portion of the triangular function of the band center frequency of the subband signal and the subband signal, and calculates a correlation function in a time zone in which the absolute value of the obtained correlation value is larger than the specified threshold value. According to the encoding apparatus of an audio signal, wherein the encoded as characteristic point information time corresponding to the maximum amplitude level and its maximum amplitude level of the band signal. Further, in the present invention, when each feature point extracting means does not extract a feature point within the maximum period in the frequency band of the subband signal, the time at which the reference level and the correlation value of the subband signal become maximum is determined. It may be encoded as feature point information. Furthermore, the feature point extracting means is
A method of encoding the signal level information of the feature point and the time difference information between the feature points extracted immediately before the feature point as a set can also be adopted.

A second invention is a receiving means for receiving, via a transmission line, data transmitted by the audio signal encoding apparatus according to the first invention, and a multiplexed data received by the receiving means for each sub-band. A data sorting unit for sorting for each signal data; and a data sorting unit that is provided corresponding to each sorting output of the data sorting unit and rearranges each feature point information of each sorting data in time series based on the time information. Feature point rearrangement means, provided in correspondence with each of the feature point rearrangement means, each inverse quantization means for inversely quantizing the rearranged feature point information, and each inverse quantization means The corresponding sub-band signal is obtained by applying an increasing or decreasing part of the trigonometric function in which the preceding and following feature points between the relocated feature points have the maximum value and the minimum value to obtain an interpolation point. Each interpolation and restoration means to be restored · The subband signal restored by the restoring means synthesizes and equipped with a signal synthesizing means for outputting a decoded audio signal according to the decoding apparatus of an audio signal characterized by.

[0008]

[Action]

Regarding the first invention, for a signal having an amplitude on the positive side from the reference level of the subband signal, the cosine function of the band center frequency of the subband signal (provided that -π / 2 ≦ angle ≦ π /
2) or a sine function (where 0 ≦ angle ≦ π) is applied,
For a signal having an amplitude on the negative side, a function in which the sign of the trigonometric function is reversed is applied, and a correlation function with the subband signal is calculated. The fact that the absolute value of the correlation value obtained from the correlation function is larger than the specified threshold value means that the correlation with the waveform of the trigonometric function is strong, and the subband signal in the time zone in which the condition is satisfied Is a feature point. That is, the maximum signal level and the corresponding time are extracted as feature point information. Therefore, even if there are a plurality of local maximums and local minimums in the section where the sub-band signal oscillates from the reference level to the positive side or the negative side, only one feature point information becomes transmission data. Significant data compression becomes possible as compared with the case where is used as a feature point. Under the above conditions, the maximum value and the minimum value other than the feature point have no auditory significance, so that the sound quality at the time of decoding / reproduction does not appear so much, and the method of setting the designated threshold value Thus, the number of feature points extracted can be controlled, and the influence on sound quality can be adjusted.

By the way, when the absolute value of the correlation value obtained from the correlation function is equal to or less than the specified threshold value, it is possible to consider that there is no feature point because the correlation with the waveform of the trigonometric function is weak. If the time zone in which no feature point exists is long, there is a possibility that a waveform faithful to the original signal may not be obtained when decoding. Therefore, when the feature point is not extracted within the maximum period in the frequency band of the sub-band signal, the time when the reference level and the correlation value of the sub-band signal become maximum is set as pseudo feature point information, so that However, the above problem is solved by ensuring the fidelity. In addition, when a feature point is represented by the maximum amplitude level information and its real time information, the data amount of the real time information becomes large. Therefore, the time information is represented by a time difference information between the feature points extracted immediately before the feature point and encoded. Is desirable.

[0010] The second invention relates to a decoding device corresponding to the encoding device of the first invention, and based on feature point information received from the encoding device via a transmission path, the original device. There is a characteristic in how the subband signal is restored. According to the present invention, based on feature point information rearranged in time series by the feature point rearranging unit and inversely quantized by the inverse quantization unit, a trigonometric function in which adjacent feature points have a maximum value and a minimum value. Connected by the interpolation curve of the increasing or decreasing part of
The interpolation points are found along the curve. Here, the “increase or decrease of the trigonometric function” is a sine function
(However, -π / 2 ≦ angle ≦ π / 2) or a cosine function (where 0 ≦ angle ≦ π). This corresponds to the fact that the feature points are obtained using the correlation function between the convex part of the trigonometric function and the subband signal in the first invention,
When feature points are continuously determined within the maximum period in the frequency band of the sub-band signal, good interpolation can be rationally realized, and even when the pseudo feature points exist, smooth interpolation is possible. Interpolation of change becomes possible.

Incidentally, the so-called "transmission path" in each invention is used not only when a signal is transmitted in real time between the transmitting means and the receiving means but also when a storage medium is interposed.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the "audio signal coding apparatus and decoding apparatus" of the present invention will be described below in detail with reference to FIGS. First, FIG. 1 shows a block circuit diagram of each device when an encoding device and a decoding device are connected via a transmission line. In the figure, an encoding device converts an input PCM signal into sub-band signals S1 to S3 of 32 bands.
A band division filter bank 1 that divides the signal into two, characteristic point extraction circuits 2-1 to 32 that extract characteristic points of the signal waveforms of the divided subband signals S1 to S32 and encode the information, A quantization circuit 3-1 to 32 for quantizing each of the quantized feature point information Sc1 to Sc32, and a multiplexer 4 for forming a bit stream by time-division multiplexing each of the quantized feature point information Dc1 to Dc32. A transmitting circuit 6 for transmitting the bit stream to the transmission path 7 side, and an input PCM signal is detected and quantization bit allocation of each of the quantization circuits 3-1 to 32 is controlled based on a psychological auditory model. It is composed of a bit allocation circuit 5.

On the other hand, the decoding device includes a receiving circuit 8 for receiving a bit stream transferred from the encoding device via a transmission path 7 and a receiving circuit 8 for converting the received bit stream to each characteristic point corresponding to each of the subbands. A demultiplexer 9 that sorts each of the information Dc1 to Dc32;
Each feature point rearrangement circuit 10-1 to 32 for rearranging each feature point based on the time information of Dc32, and each inverse quantization circuit 11-1 to 32 for inversely quantizing each rearranged feature point information And interpolation / restoration circuits for creating sub-band signals S'1 to S'32 by interpolating and restoring between feature points based on the dequantized feature point information.
A synthesis filter bank 13 for synthesizing 12-1 to 32 and the restored sub-band signals S'1 to S'32 to reproduce the original PCM signal
It is composed of

The above circuit configuration is substantially the same as that of a conventional image signal encoding and decoding apparatus.
In this embodiment, each feature point extracting circuit 2-1 to 3 on the encoding device side is used.
There is a feature in the feature point extraction method in 2, and the interpolation method in each interpolation and restoration circuit 12-1 to 32 on the decoding device side,
Hereinafter, those systems will be specifically described for the sub-band signal having the waveform shown in FIG.

First, each feature point extraction circuit 2-
The subband signals S1 to S32 divided by the band division filter bank 1 are input to 1 to 32, and these signals S1 to S32
Are 0 ~ (1/32) ω, (1/32) ω ~ (2/32) ω, ..., (31/3
2) ω to ω: [where ω is the angular frequency] relates to the frequency band. Each of the feature point extraction circuits 2-1 to 32 has a cosine function of b1 (t) = cos (.theta.t + .tau.) And b2 (t) =-cos (.theta.t + .tau.). Here, θ corresponds to the central angular frequency of the frequency band of the corresponding subband signals S1 to S32,
θt is in the range of −π / 2 ≦ θt ≦ π / 2, and the cosine functions: b1 (t) and b2 (t) are defined as positive and negative convex functions, respectively. Τ is a variable for moving the cosine function in the time axis direction, and indicates a time difference between the subband signal and the cosine function.

Now, as shown by a solid line in FIG. 2, a subband signal (a signal waveform similar to that in FIG. 16) represented by a time function: a (t) is input to the feature point extraction circuit 2-n. Case 3
As shown in the figure, the feature point extraction circuit 2-n uses the above cosine functions: b1 (t) and b2 (t) to calculate a correlation function: R1ab (τ) =
Let ∫a (t) · b1 (t) dt and R2ab (τ) = ∫a (t) · b2 (t) dt be [−
演算 <t <∞] (actually operated on the length of the subband signal), and while changing τ, the subband signal:
A correlation value between a (t) and each cosine function: b1 (t), b2 (t) is obtained. In this case, each correlation function using the time difference τ as a variable: R1
ab (τ) and R2ab (τ) change in a substantially triangular waveform corresponding to the amplitude of the sub-band signal: a (t) as shown in FIG.

The feature point extraction circuit 2-n is provided with threshold values Th and -Th for judging the magnitude of the correlation value in advance, and the correlation value is determined by each of the threshold values Th and -Th. The maximum value or the minimum value of the sub-band signal: a (t) in a time zone exceeding? That is, the above cosine function:
By observing the correlation between b1 (t) and b2 (t) and the amplitude waveform of the sub-band signal: a (t), one feature point is obtained on condition that a certain degree of correlation is obtained. I have to. Therefore, when each feature point of the sub-band signal: a (t) is obtained by the algorithm, the sub-band signal: a (t) is obtained as shown in FIG.
Extreme values of: Pn, Pn + 1, Pn + 3, Pn + 4 are obtained as feature points,
The feature point extraction circuit 2-n determines the signal level and time of those feature points as feature point information: (Hn, tn), (Hn + 1, tn + 1), (Hn + 3, tn +
3), (Hn + 4, tn + 4).

Incidentally, the sub-band signal: a (t) in FIG. 2 has amplitudes on the positive side and the negative side within the maximum period of the signal even after time zones: tn + 1 to t + 3 and tn + 4. However, as shown in FIG. 4, since the absolute value of the correlation value is smaller than the threshold value Th, no feature point is obtained. In such a case, processing may be performed assuming that there is no feature point, but in this embodiment, the time at which the correlation value is maximum in each time zone: tn + 2, tn
The +5 and 0 levels are set as pseudo feature points: Pn + 2, Pn + 5. Therefore, the feature point information is (Hn, tn), (Hn + 1, tn + 1),
(0, tn + 2), (Hn + 3, tn + 3), (Hn + 4, tn + 4), (0, tn + 5)
Becomes Further, if each time information is used as it is, the redundancy increases. Therefore, each time information is actually expressed as a time difference from the immediately preceding feature point, and the feature point information is represented by (Hn, tn−tn−
1), (Hn + 1, tn + 1-tn), (0, tn + 2-tn + 1), (Hn + 3, tn
+ 3-tn + 2), (Hn + 4, tn + 4-tn + 3), (0, tn + 5-tn + 4)
And encodes the respective feature point information.

Each of the encoded feature point information is quantized by a quantization circuit 3-n, and a multiplexer 4 forms a bit stream time-division multiplexed with feature point information on other subband signals. The signal is transmitted from the transmission circuit 6 to the transmission path 7. Here, for the same subband signal, the feature point of FIG. 2 obtained by the algorithm of the present embodiment is compared with the feature point of FIG. 16 obtained by the conventional algorithm using the extremum as a feature point. In FIG. 16, only the information of the five feature points Pn to Pn + 5 is to be transmitted, whereas in the case of FIG. 16, all the extrema occurring at a small amplitude of the signal are the feature points Pn to Pn + 11. Therefore, it is understood that significant data compression can be realized by the encoding device of this embodiment. In addition, since a change in the sub-band signal at a small amplitude does not have a significant auditory significance, practically no deterioration in reproduced sound quality occurs.

Next, in the encoding apparatus shown in FIG.
How the (PCM signal) is processed will be described with reference to an actual signal waveform. First, when the original signal has a signal waveform as shown in FIG. 5, each sub-band signal divided by the band division filter bank 1 is shown in FIG. 6 to FIG.
The signal waveform is as shown in FIG. However, here, for example, the band is (31/32) ω to ω, (20/32) ω to (21/32)
ω, (10/32) ω to (11/32) ω, and subband signals of 0 to (1/32) ω are shown.

The feature points extracted by the feature point extraction circuits 2-32, 2-21, 2-11 and 2-1 corresponding to the subbands are shown in FIGS. 6 to 9B. The point is as shown. As is clear from the state of extraction of the feature points, when the sub-band signal to be extracted has a high frequency band, a fine amplitude is included, and when the feature points are extracted with extreme values, an extremely large number of feature points are generated. However, as can be seen in FIGS. 6 and 7, the feature point extraction circuit 2-n of the present embodiment extracts only the maximum value and the minimum value as feature points in macroscopic view of the amplitude. The number has been greatly reduced.

On the other hand, in a decoding apparatus which receives a bit stream in which each feature point information is coded and multiplexed for each sub-band via a transmission path 7, a signal obtained from a reception circuit 8 is sub-multiplexed by a demultiplexer 9. Feature point information Dc1 for each band
Dc32, the feature points are rearranged by the feature point rearrangement circuits 10-1 to 32, and dequantized by the inverse quantization circuits 11-1 to 32. Then, the interpolation / restoration circuits 12-1 to 32 are performed. With subband signals S'1 to S'32
To restore.

In this case, the algorithm in the interpolation / restoration circuit 12-n is obtained on the encoding device side at each feature point P in FIG.
An example in which the invention is applied to n to Pn + 5 will be described. First,
Each feature point Pn, Pn + rearranged as shown in FIG.
The following cosine function between 1, Pn + 2, Pn + 3, Pn + 4, Pn + 5,
Connect with the curve defined by Yx = + (Ax / 2) · cos {(π / Tx) · [t− (Tx / 2)]} + (1/2) · (Hx−Hx + 1) where [0 ≦ (π / Tn) · t ≦ π, x = n, n + 2, n + 4] Yx = − (Ax / 2) · cos {(π / Tx) · [t− (Tx / 2)]} + (1 / 2) · (Hx−Hx + 1) where [0 ≦ (π / Tn) · t ≦ π, x = n + 1, n + 3] That is, ± where each feature point has a maximum value and a minimum value Prediction approximation is smoothly performed between each feature point using a cosine function. In that case,
Since the encoding apparatus uses the correlation function between the subband signal and the cosine function to find the feature points, a reasonable interpolation curve can be obtained.

Then, the sub-band signals S'1 to S'32 are restored by obtaining interpolation values at regular intervals based on the interpolation curves formed between the characteristic points Pn to Pn + 5. . Here, the sub-band signals S′1 to S′32 restored by the respective interpolation / restoration circuits 12-1 to 32 are restored from the feature points discretely extracted by the algorithm on the encoding device side. For this reason, the waveforms of the sub-band signals S1 to S32 divided by the band division filter 1 on the encoding device side do not have the same waveform, but all the waveform portions having a high auditory importance are restored. Therefore, when the reconstructed sub-band signals S′1 to S′32 are synthesized by the synthesis filter bank 13, a PC that has almost the same acoustic properties as the PCM signal input to the encoding device side
An M signal is obtained.

For example, when the sub-band signal is restored based on the characteristic points obtained in the extraction state as shown in FIGS. 6 to 9B, the sub-band signals are restored as shown in FIGS. 11 to 14, respectively. The decoded signal (PCM signal) after being synthesized by the synthesis filter bank 13 has a signal waveform as shown in FIG. Here, the waveform of the original signal in FIG.
Comparing the waveforms of the decoded signals of FIG. 5, the decoded signal of FIG. 15 shows that the amplitude of the original signal at the fine level is lost, but all the signal waveforms of a large amplitude that are perceptually important are restored. It is understood that there is. Therefore, according to this embodiment, when the encoding device transmits the audio information to the decoding device with an extremely small amount of data and reproduces the output of the decoding device, the audio signal input to the encoding device is reproduced. Can be reproduced with almost the same sound quality as when directly reproduced.

[0026]

According to the "audio signal encoding apparatus and decoding apparatus" of the present invention, the following effects are obtained by having the above configuration. According to a first aspect of the present invention, in a coding apparatus that divides an audio signal into subband signals and uses feature point information of each subband signal as transmission information, a conventional method of extracting maximum / minimum values as feature points is as follows. The number of feature points could not be applied to the audio signal because the number of feature points was too large, but the feature points were extracted using the correlation function of the subband signal and the convex part of the trigonometric function. It is possible to transmit information with high auditory importance while greatly reducing the amount of data, and to improve transmission efficiency and reduce the capacity of a storage medium by compressing data with almost no deterioration in sound quality. Further, by changing the threshold value used for extracting the feature points, the fidelity of the decoded signal with respect to the original signal can be adjusted, so that it can be easily adapted to the specifications of the audio system. According to a second aspect of the present invention, when a feature point is not extracted within a maximum period in a band of a subband signal,
A feature point that is as close as possible to the original signal is set, so that a more faithful reconstruction of the subband signal is enabled on the decoding device side. The third aspect of the present invention reduces the redundancy of the time information in the feature point information, thereby further improving the transmission efficiency and saving the capacity of the storage medium. The invention of claim 4 is
It is used in correspondence with the encoding apparatus of each of the above-described inventions, and enables a rational and faithful restoration of a subband signal by interpolating between feature points with a curve of an increasing portion or a decreasing portion of a trigonometric function. .

[Brief description of the drawings]

FIG. 1 is an overall block circuit diagram in a case where an encoding apparatus and a decoding apparatus according to an embodiment of an “audio signal encoding apparatus and decoding apparatus” of the present invention are connected via a transmission path. .

FIG. 2 is a graph showing a signal waveform of a subband signal, characteristic points obtained by applying a correlation function between the signal and a cosine function, and pseudo characteristic points when characteristic points cannot be obtained.

FIG. 3 is a graph showing an application mode of a correlation function.

FIG. 4 is a graph showing a change state of a correlation value obtained by applying a correlation function to a subband signal.

FIG. 5 is a graph showing a signal waveform of an original signal (PCM signal) input to the encoding device.

FIG. 6 is a graph showing a signal waveform (A) of a sub-band signal of a band: (31/32) ω to ω, and an extraction mode (B) of a feature point thereof.

FIG. 7 is a graph showing a signal waveform (A) of a sub-band signal of a band: (20/32) ω to (21/32) ω, and an extraction mode (B) of a feature point thereof.

FIG. 8 is a graph showing a signal waveform (A) of a sub-band signal having a band of (10/32) ω to (11/32) ω, and an extraction mode (B) of a feature point thereof.

FIG. 9 is a graph showing a signal waveform (A) of a subband signal having a band of 0 to (1/32) ω and an extraction mode (B) of a feature point thereof.

FIG. 10 is a graph showing an interpolation algorithm between each feature point in the interpolation / restoration circuit on the decoding device side.

FIG. 11 is a graph showing a signal waveform of a subband signal of a restored band: (31/32) ω to ω.

FIG. 12 is a graph showing signal waveforms of subband signals in a restored band: (20/32) ω to (21/32) ω.

FIG. 13 is a graph showing signal waveforms of subband signals in a restored band: (10/32) ω to (11/32) ω.

FIG. 14 is a graph showing signal waveforms of subband signals in the restored band: 0 to (1/32) ω.

FIG. 15 is a graph showing a signal waveform of a decoded signal obtained by combining the restored subband signals.

FIG. 16 is a graph showing a manner of generating feature points in a conventional method for extracting a minimum value and a maximum value of a subband signal as feature points.

[Explanation of symbols]

1 ... band division filter bank (band division means), 2-1 to 32 ...
Feature point extraction circuit (feature point extraction means), 3-1 to 32: quantization circuit (quantization means), 4: multiplexer (transmission means), 5: bit allocation circuit, 6: transmission circuit (transmission means), 7 … Transmission line,
8 reception circuit (reception means), 9 demultiplexer (data classification means), 10-1 to 32 ... feature point rearrangement circuit (feature point rearrangement means), 11-1 to 32 ... inverse quantization circuit (inverse Quantization means), 12-1
.About.32: interpolation / restoration circuit (interpolation / restoration means), 13: synthesis filter bank (signal synthesis means).

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) H03M 7/30

Claims (4)

(57) [Claims] 1. A band dividing unit for dividing an input audio signal into a plurality of sub-band signals, and a feature point provided for each divided output of the band dividing unit, wherein a characteristic point is determined from a signal waveform of the divided sub-band signal. Each feature point extracting means for extracting and obtaining signal level information and time information of each feature point is provided corresponding to each of the feature point extracting means, and the feature point information obtained by each of the feature point extracting means is provided. In an audio signal encoding apparatus comprising: each quantizing means for quantizing; and a transmitting means for multiplexing and transmitting the quantized data by each of the quantizing means, each of the feature point extracting means includes a subband signal. Calculates the correlation function between the convex part of the trigonometric function of the band center frequency of the band and the sub-band signal, and calculates the maximum amplitude level of the sub-band signal in the time zone where the absolute value of the obtained correlation value is larger than the specified threshold An encoding apparatus for an audio signal, which encodes a time corresponding to the maximum amplitude level as feature point information. 2. When each feature point extracting means does not extract a feature point within the maximum period in the frequency band of the subband signal, a time at which the reference level and the correlation value of the subband signal become maximum is set as the feature point. 2. The audio signal encoding apparatus according to claim 1, wherein the encoding is performed as information. 3. The method according to claim 1, wherein each feature point extracting means encodes a set of signal level information of the feature point and time difference information between feature points extracted immediately before the feature point. The audio signal encoding device according to claim 2. 4. A receiving means for receiving, via a transmission path, data transmitted by the audio signal encoding apparatus according to claim 1, and multiplexed data received by the receiving means. Data sorting means for sorting each data of the sub-band signal, and provided in correspondence with each sorting output of the data sorting means, and each feature point information of each sorting data is rearranged in chronological order based on the time information. Each feature point rearrangement means to be provided, provided corresponding to each of the feature point rearrangement means,
Each inverse quantization means for inversely quantizing each rearranged feature point information, and each inverse quantization means are provided corresponding to each inverse quantization means, and the maximum number of preceding and following feature points between each rearranged feature point is maximum. Each interpolation / restoration means for restoring a sub-band signal by calculating an interpolation point by applying an increasing portion or a decreasing portion of a trigonometric function which becomes a value and a minimum value, and a sub-band signal restored by each interpolation / restoration means. An audio signal decoding device comprising a signal synthesizing means for synthesizing and outputting a decoded audio signal.