JPH0535300A - Voice recording and reproducing device - Google Patents

Abstract

PURPOSE:To prevent the generation of nonlinear noises which are generated at the time of phrase completion during a reproduction. CONSTITUTION:The device is provided with a low pass filter 1 which eliminates foldover noises of analog voice signals, an A/D converter 2 which converts these analog voice signals to PCM signals, a companding processing circuit 3 which compands the PCM signals, a RAM 4 which stores the compressed data, a D/A converter 6 which converts the PCM signals to analog signals, a control circuit 8 which controls the system and a soft muting circuit 5 which detects the completion time of the voice generation during the voice signal reproduction and gradually attenuates prior to the completion.

Description

Detailed Description of the Invention

[0001]

The present invention relates to a PCM (Pulse Code Mod)
modulation signal and compressing and decompressing the same, and storing the compressed data in RAM. More particularly, the present invention relates to a voice recording and reproducing apparatus having a muting function.

[0002]

2. Description of the Related Art Conventionally, in a voice recording / reproducing apparatus, an analog signal is converted into a PCM signal and ADPCM (Adapti
Audio data is compressed by a data compression method such as ve Differential PCM method and stored in RAM. At this time, the recording is performed for each phrase, and the time for one phrase is set to F seconds, and when the recording end time of the phrase is reached, the data compression processing is unconditionally stopped and the recording is ended. For this reason, the above-mentioned RAM stores the compressed data whose compression processing has ended halfway.

Next, the compressed data in the RAM is expanded to convert the compressed data into PCM data. An analog signal can be obtained by DA converting the PCM data. At this time, since the compressed data is converted into the PCM data and the data value when reaching the end data becomes zero, nonlinear noise is output. As a result, the reproduced audio signal converted into the analog signal generated non-linear noise due to a sudden interruption of the audio signal at the end of reproduction.

[0004]

Therefore, according to the conventional technique, since the compression processing is performed within a fixed time and the compressed data that has been compressed is expanded within the fixed time, non-linear data is generated when the data disappears. As a result, there is a problem that non-linear noise occurs and the sound quality deteriorates.
Therefore, the present invention has been made in consideration of the above circumstances,
It is an object of the invention to provide a voice recording / reproducing apparatus which prevents non-linear noise generated at the end of a phrase during reproduction.

[0005]

In order to achieve the above object, in a voice recording / reproducing apparatus according to the present invention, a low-pass filter for removing aliasing noise of an analog voice signal and a PCM for this analog voice signal. A / D converter for converting into a signal, a compression / expansion processing circuit for compressing / expanding the PCM signal, and a RAM for storing the compressed data
And a DA converter for converting the PCM signal into an analog signal, a control circuit for controlling the system, and a soft muting circuit for detecting a sound generation end time at the time of reproducing the sound signal and gradually attenuating before the end. It is characterized by that.

[0006]

In the present invention having the above structure, even if the compressed data is interrupted during the data compression process and the compressed data remains non-linear, the soft muting circuit performs the decompression process from PCM before T seconds before the final address of the compressed data. Soft muting the PCM data by multiplying the data by an attenuation factor. As a result, the audio data is gradually attenuated before the end of the reproduction signal, and the audio output is prevented from becoming non-linear before the end.

[0007]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 shows an embodiment of a voice recording / reproducing apparatus according to the present invention. As shown in the figure, an analog audio signal input from a microphone amplifier is processed by a low-pass filter (hereinafter referred to as L
It is called PF. ) 1 limits the band to attenuate noise such as aliasing noise. The analog audio signal from the LPF 1 is converted into a PCM signal of about 10 to 12 bits by the AD converter 2. The sampling clock of the AD converter 2 is supplied from the control circuit 8. AD converter 2
The PCM signal output from is supplied to the compression / expansion processing circuit 3 where it is compressed. The compression / expansion processing circuit 3 is a circuit block that performs encoding / decoding processing such as ADPCM. The algorithm of the compression / expansion processing circuit 3 is not particularly specified, but in the present embodiment, 4-bit ADPCM is used. PCM signal is 4-bit ADPC
It is compressed to M and stored in the RAM 4. This compression process is 1
Perform a phrase (F seconds) and store one phrase in the RAM 4 to finish. Compression / expansion processing circuit 3 for this compression, RA
The control circuit 8 controls M4. This is a compression process, that is, a voice recording process.

Next, an audio reproduction process which is a decompression process will be described. 4-bit ADP stored in RAM4
The CM data is compressed / expanded by the PCM signal (10
Decompression processing to about 12 bits). By supplying the PCM signal to the DA converter 6, the PCM signal is converted into an analog audio signal. Then, by supplying the analog signal to the LPF 7, an analog audio signal with reduced aliasing noise is output. This analog voice signal is sounded as voice through a power amplifier and a speaker. At this time, the control circuit 8 controls the compression / expansion processing circuit 3, the soft muting circuit 5 and the sampling clock of the DA converter 6.

In this embodiment, the soft muting circuit 5 operates to reduce the non-linear noise generated at the end of the phrase. This circuit 5 detects the end of a phrase and applies soft muting before the end. That is, at the time of decompression processing, the soft muting circuit 5 performs PCM from T seconds before the final address of the compressed data.
Soft muting the PCM data by multiplying the data by an attenuation factor. Hereinafter, FIG. 2 and FIG.
This soft muting will be explained using.
At the time of compression processing, the start address and the end address of the phrase are stored in the RAM 4. This final address is stored in the final value address register 11 via the control circuit 8. The mute start address generation circuit 12 generates a mute start address based on the value of the final value address register 11. This generation method is an address obtained by subtracting the soft muting time T seconds from the final address. From this soft muting time T seconds to the sampling period Ts
By dividing the second, the address value of the soft muting time can be obtained.

The address of the RAM 4 during phrase reproduction is supplied to the comparison circuit 13 via the control circuit 8. The comparison circuit 13 compares the RAM address with the soft muting start address, and if they match, supplies a match signal to the attenuator value output circuit 14. Upon receiving this signal, the attenuator value output circuit 14 generates a signal that is sequentially attenuated according to the sampling cycle. Next, in the attenuator circuit 15, the soft muting signal output from the attenuator value output circuit 14 and the compression / expansion processing circuit 3
The PCM signal output from is multiplied. By this processing, soft muting processing can be performed on the PCM signal. This PCM signal is sent to the DA converter 6, LPF
It is output as an analog audio signal via 7. By this processing, it is possible to configure a voice recording / reproducing apparatus in which nonlinear noise is reduced at the end of the phrase.

[0011]

As described above, according to the present invention,
Since the non-linear noise generated at the end of the phrase at the time of reproduction does not occur, it is possible to provide a voice recording / reproducing apparatus with higher sound quality. Then, it is possible to generate a synthetic sound that is easy to hear.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of a voice recording / reproducing apparatus according to the present invention.

FIG. 2 is a block diagram showing a soft muting circuit in this embodiment.

FIG. 3 is an explanatory diagram showing soft muting in the present embodiment.

[Explanation of symbols]

 1 LPF 2 AD converter 3 compression / expansion processing circuit 4 RAM 5 soft muting circuit 6 DA converter 7 LPF 8 control circuit 11 final value address register 12 mute start address generation circuit 13 comparison circuit 14 attenuator value output circuit 15 attenuator circuit

Claims (1)

Claim: What is claimed is: 1. A low-pass filter for removing aliasing noise of an analog audio signal, and P
An AD converter for converting into a CM signal, a compression / expansion processing circuit for compressing / expanding the PCM signal, a RAM for storing the compressed data, a DA converter for converting the PCM signal into an analog signal, and control for controlling the system. An audio recording / reproducing apparatus comprising: a circuit; and a soft muting circuit that detects an audio generation end time when reproducing an audio signal and attenuates gradually before the end.