JPS61131032A - Voice data accumulating system - Google Patents

Abstract

PURPOSE:To manage voice data accumulation with a memory having a small capacity, by making ADPCM-coded voice data blocks invalid when the number of cases where the level of the voice data of a PCM-coding circuit exceeds a reference level is lower than a reference number of times. CONSTITUTION:Voice signals supplied to an input IN are changed to ADPCM codes by means of an ADPCM coding circuit 2 after they are changed to PCM codes at a PCM coding circuit 1 and temporarily accumulated in a data buffer 4. The output of the PCM codes is compared with a reference level Lv given from a level setting circuit 6 at a level comparator circuit 3 and the number of cases where the voice data exceeds the reference level Lv is counted by means of a counter circuit 5 which inputs the output of the comparator circuit 3. A microprocessor 7 stores ADPCM voice data blocks in the data buffer 4 in a memory 8 only when the count value exceeds a reference number of times.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to an audio data storage method, and in particular, to
The present invention relates to an audio data storage method that encodes CM, further encodes ADPCM, and stores the encoded data in a memory.

The audio data storage method that encodes and stores the audio signal requires a huge amount of memory. It was strongly desired by

[Conventional technology]

Conventional audio data storage methods have adopted a method in which an audio signal is PCM encoded, and the PCM encoded audio signal is further ADPCM encoded and stored.

The reason why the above method is adopted is that the PCM code has 8 bits including the polarity, but the ADPCM code only needs 4 bits, which has the advantage of halving the memory capacity.

[Problem that the invention seeks to solve]

SUMMARY OF THE INVENTION An object of the present invention is to provide an audio data storage method that requires a smaller memory capacity than the conventional method described above.

[Means for solving problems]

The means to solve the problem is the PCM encoding circuit and A
A DPCM encoding circuit is provided, the PCM encoding circuit and the A
In an audio data storage device that encodes an audio signal using a DPCM encoding circuit and stores it in a memory, the ADPCM
Divide the audio data encoded by the encoding circuit into blocks of multiple bytes, compare the level of the audio data encoded by the PCM encoding circuit with a reference level, and count the number of times the audio data exceeds the reference level. When the counted value is greater than or equal to a certain reference number of times, the divided audio data block is determined to be valid, and when it is less than or equal to the reference number of times, the divided audio data block is determined to be invalid and is determined to be valid. This is achieved by storing only the audio data block and the audio data blocks before and after it.

[Effect]

According to the present invention, a PCM encoded audio signal is divided into blocks of multiple bytes, and the number of signals above a reference level is counted in this block. If it is above, the block is stored as valid, and if it is below, it is discarded as invalid, so only the truly valid data in the audio signal is stored, so it requires much less memory capacity than the conventional method. An effect is produced.

〔Example〕

FIG. 1 is a diagram showing an embodiment of the audio data storage method according to the present invention.

FIG. 2 is an explanatory diagram of the operation of FIG. 1.

In the figure, 1 is a PCM encoding circuit, 2 is an ADPCM encoding circuit, 3 is a level comparison circuit, 4 is a data buffer, 5 is a counting circuit, 6 is a level setting circuit, 7 is a microprocessor, and 8 is a memory.

■ The present invention will be described in detail below with reference to the drawings.

The audio signal enters the PCM encoding circuit 1 and is converted from A to D. In this case, it is normal to convert to 8 bits including the polarity bit.

Here, a plurality of bytes m is defined as one block.

The audio signal PCM encoded by the PCM encoding circuit 1 is compared with a reference level Lv in the level comparison circuit 3 every time it is encoded. This reference level Lv (absolute value) is supplied from the level setting circuit 6.

The level comparison circuit 3 outputs an output pulse when audio data higher than the reference level Lv is input.

The counting circuit 5 is reset every block and counts the number of output pulses in one block (m bytes of data).

Also, the audio signal PCM encoded by the PCM encoding circuit 1 is the same as the conventional method (enters the ADPCM encoding circuit 2,
The data is ADPCM encoded and temporarily stored in the data buffer 4.

The microprocessor 7 reads the count value of the counting circuit 5 at the time before the reset, and when this count value is greater than the reference value N, it determines that the audio data of the block is valid and stores the contents of the data buffer 4 in the memory 8. Store. If the count value is the standard (
When it is smaller than tiN, it is determined that the audio data of the block is invalid and the contents of the data buffer 4 are discarded.

FIG. 2(a) illustrates the above explanation, and Lv is the reference level. The audio signal (P
CM code) is input to the level comparison circuit 3, 5(0)
When , the level comparison circuit 3 is lower than the reference level Lv.
does not output an output pulse, and when 5 (1), the reference level Lv
Since the above is the case, the level comparison circuit 3 outputs an output pulse.

In this manner, assuming that the waveform shown in FIG. 2(a) is one block, output pulses are generated seven times within one block (S (01 to 5 (9)).

The counting circuit 5 counts this, and the microprocessor 7 makes it valid when the count value 7 is larger than the reference value N, and invalidated when it is smaller than N.

FIG. 2(b) is a diagram illustrating a case where the processing of the present invention is applied to a series of audio signals.

By taking the measures described above, silent parts included in the audio signal and parts close to this can be deleted, and only truly effective audio signals can be stored.

In this case, when valid blocks are continuously detected, it is also possible to increase the fidelity by considering invalid blocks before and after the valid blocks to be particularly valid and storing them.

〔Effect of the invention〕

As described above in detail, the present invention has the great effect of realizing an audio data storage system that can store audio signals with significantly less memory than conventional systems.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing an embodiment of the audio data storage method according to the present invention. FIG. 2 is an explanatory diagram of the operation of FIG. 1. In the figure, 1 is a PCM encoding circuit, 2 is an ADPCM encoding circuit, 3 is a level comparison circuit, 4 is a data buffer, 5 is a counting circuit, 6 is a level setting circuit, 7 is a microprocessor, and 8 is a memory.

Claims (1)

[Claims] The PCM encoding circuit and the ADPCM encoding circuit are provided.
In an audio data storage device that encodes an audio signal using a CM encoding circuit and the ADPCM encoding circuit and stores it in a memory, the audio data encoded by the ADPCM encoding circuit is divided into blocks of multiple bytes, Comparing the level of the audio data encoded by the PCM encoding circuit with a reference level and counting the number of times the level exceeds the reference level;
When the counted value is greater than or equal to a certain reference number of times, the divided audio data block is determined to be valid, and when it is less than or equal to the reference number of times, the divided audio data block is determined to be invalid and determined to be valid. An audio data storage method characterized in that only the audio data block and the audio data blocks before and after the audio data block are stored.