JPH0744199A - Speech sound recording and reproducing device - Google Patents

Abstract

PURPOSE:To prevent break sound which is offensive to the ear from being generated at the time of reproduction by forcibly recording soundless data for a certain period of time after the sound recording is started and before the sound recording is ended. CONSTITUTION:This device is equipped with a soundless data recording means which records the soundless data for a certain period of time (muting time) after the sound recording is started and before the sound recording is ended and the soundless data recording means consists of a digital signal processing circuit 6 and a control part 9. Then the soundless data generated by the digital signal processing circuit 6 ar recorded while a certain period of muting time is set. Consequently, neither the mechanical noise of a sound recording switch nor the bias variation noise of a microphone, etc., generated at the time of start and end of the sound recording is recorded in a semiconductor memory 7. Further, a break sound due to a noise is eliminated to be outputed from a speaker 3 at the time of reproduction. Further, the muting time can accurately be limited to a minimum and speech cutting due to an unnecessary long muting time is precluded.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice recording / reproducing apparatus which uses a semiconductor memory as a recording medium.

[0002]

2. Description of the Related Art A voice recording / reproducing apparatus of this kind converts a voice analog signal input at the time of recording into a digital signal by an A / D converter, converts the digital signal into frames at a constant time interval, and outputs each frame unit. The compressed data is recorded in a semiconductor memory, the compressed data read from the semiconductor memory at the time of reproduction is expanded for each frame, and the expanded digital signal is converted into an audio analog signal by a D / A converter. It was designed to output from the speaker.

By the way, in such a voice recording / reproducing apparatus, mechanical noise of a recording switch and bias fluctuation noise of a microphone, an amplifier, etc. are generated at the start of recording and at the end of recording. As a result, the noise is recorded in the memory, and as a result, the noise is output as a clicking sound at the time of reproduction, which is annoying.

Therefore, in order to solve such a problem, a hardware muting circuit is added to an analog recording circuit system such as a microphone and an amplifier.

[0005]

However, the above-mentioned conventional device requires a hardware muting circuit, and the muting time cannot be set very finely. The voice data was cut more than necessary, and it could not be said that it was very preferable.

However, according to the present invention, a sound recording / reproducing apparatus capable of preventing an audible pop noise at the time of reproduction due to noise and suppressing the cut audio data to a minimum without using a hardware muting circuit. I will provide a.

[0007]

According to a first aspect of the present invention, there is provided a voice recording / reproducing apparatus for converting an input voice analog signal into a digital signal, converting the digital signal into frames at a constant time interval, and compressing each frame unit. Then, the compressed data is recorded in a semiconductor memory, and at the time of reproduction, the compressed data read from the semiconductor memory is expanded frame by frame, and the expanded digital signal is converted into an audio analog signal and output. For semiconductor memory,
A silent data recording means is provided for recording the silent data for a fixed time after the start of recording and for a fixed time before the end of recording.

Further, in the audio recording / reproducing apparatus according to the second aspect of the present invention, the input audio analog signal is converted into a digital signal, the digital signal is framed at a constant time interval, and compressed for each frame unit. Data is recorded in a semiconductor memory, and at the time of reproduction, compressed data read from the semiconductor memory is expanded frame by frame, and the expanded digital signal is converted into an audio analog signal and output. There is provided reproduction control means for reproducing the data of each recording block recorded in the memory from the beginning except for a fixed time and a fixed time before the end.

Further, in the voice recording / reproducing apparatus according to the third aspect, in addition to the configuration according to the second aspect, a reproduction start marker is provided in the semiconductor memory at an address position after a predetermined time from the start of the recording, and the recording end is fixed. The reproduction control marker recording means for recording the reproduction end marker at the address position before the time is provided, and the reproduction control means detects the reproduction start marker to start the reproduction, and ends the reproduction by detecting the reproduction end marker.

[0010]

According to the structure of the above-mentioned claim 1, by forcibly recording the silent data after the start of recording and before the end of recording (muting time), the mechanical noise of the recording switch, the microphone, When the bias fluctuation noise of the amplifier or the like is generated, these noises are not recorded in the semiconductor memory, and no jarring pop noise is generated during reproduction. In addition, the muting time can be set finely and accurately with the length of one frame as the minimum time, which causes an inconvenience that the muting time becomes longer than necessary and the audio data to be originally recorded is cut off accordingly. Also disappears.

According to the second and third aspects of the invention, the data of each recording block is reproduced from the beginning except for a fixed time and a fixed time before the end, and mechanical noise and bias are reproduced. When the fluctuating noise is recorded, those noises are not reproduced at the time of reproduction, so that no jarring pop noise is generated. Moreover,
Since the reproduction start / end positions can be set finely and accurately according to the time of occurrence of noise, there is no inconvenience that the audio data is cut more than necessary.

[0012]

Embodiments of the present invention shown in the drawings will be described in detail below. FIG. 1 is a block diagram showing a voice recording / reproducing apparatus according to an embodiment of the present invention. In the figure, 1 is a microphone, 2 is an amplifier, 3 is a speaker, 4 is an amplifier, and 5 is an analog signal converted into a digital signal. A / D conversion function and D / A for converting digital signals to analog signals
A converter having a conversion function, 6 is a digital signal processing circuit having a function of converting a digital signal into a frame and performing compression processing in frame units, a function of expanding compressed data in each frame, and the like, 7 is a semiconductor memory such as an IC card,
Reference numeral 8 is a switch group, 9 is a control unit for controlling the above-mentioned respective units, and the control unit 9 is mainly composed of a microcomputer.

Thus, the present embodiment is provided with a silent data recording means for forcibly recording the silent data after the start of recording and before the end of recording (muting time). The silence data recording means is composed of the digital signal processing circuit 6 and the control unit 9.

FIG. 3 is a conceptual diagram showing a data recording state in a semiconductor memory. Point A indicates a recording start position and point B indicates a recording end position. In the illustrated example, point A to AS.
To 5 points each, from BE point to B point,
It records silent data, and if the length of one frame is 20 milliseconds, the muting time is 100 milliseconds.

Next, the operation at the time of recording will be described with reference to the conceptual diagram showing the data recording state of FIG. 3 and the control flowchart of FIG. Now turn on the recording switch
Then, the control unit 9 determines that the recording switch is ON, sets the recording mode, and activates the converter 5 and the digital signal processing circuit 6, while the voice is converted into an electric signal (analog signal) by the microphone 1. , Is input to the converter 5 via the amplifier 2, converted into a digital signal, and then input to the digital signal processing circuit 6, and the control unit 9 controls according to the control flowchart of FIG.

In the recording mode state, the control unit 9 first sets the value MF of the number of muting frames to "5", and then sets the frame count number FS to "1". Next, it is confirmed that the recording switch is ON, and the muting frame number MF and the frame count number FS are compared. If MF ≧ FS, the digital signal processing circuit 6
At S1, the silent data is generated, the silent data is written to the semiconductor memory 7 by specifying an address, and the frame count number FS is incremented by 1.

Then, such silent data writing is repeated until MF <FS. That is, as shown in FIG. 2, the silence data is recorded for 5 frames from the point A. Therefore, the mechanical noise of the recording switch and the bias fluctuation noise of the microphone 1 and the amplifier 2 which are generated during this period are not recorded.

Next, when the frame count number FS becomes "6" and the relationship of MF <FS is satisfied, the voice data converted into the digital signal by the converter 5, that is, the voiced data is compressed, and then the semiconductor memory 7 is used. The address is designated and written to, and the frame count number is incremented by "1". After that, the recording switch turns off such operation.
It is repeated for each frame until F is reached, and voiced data is recorded after 6 frames of the semiconductor memory 7 as shown in FIG.

When the recording switch is turned off,
The control unit 9 determines whether the recording switch is OFF,
S is compared, and if MF ≧ FS, the recording mode state is released.

On the other hand, if MF ≧ FS is not satisfied, the muting frame number MF is subtracted from the frame count number FS at that time to obtain the frame count number FE at the BE point in FIG.
The address corresponding to the above FE (BE) frame is designated, the silent data generated by the digital signal processing circuit 6 is written (overwritten), and the frame count number F
E is incremented by "1". After that, while the relationship of FS ≧ FE is satisfied, the writing of the silent data is continued, and FS <F
When the relationship of E is established, the recording mode state is released.

Therefore, as shown in FIG. 2, although voiced data is once written in the 5 frames before the point A at the end of recording, it is forcibly rewritten to silent data by overwriting, and as a result, The above mechanical noise and bias fluctuation noise generated during this period are not recorded and remain.

As described above, in the present embodiment, the constant muting time after the start of recording and before the end of recording is set, and during this period, the silent data generated by the digital signal processing circuit 6 is recorded. The mechanical noise of the recording switch and the bias fluctuation noise of the microphone generated at the start and end of the recording are not recorded in the semiconductor memory 7. Therefore, the clicking noise caused by the noise is output from the speaker 3 during the reproduction. It will not be done.

Further, since the muting time can be finely set with the length of one frame as a minimum time, the period during which noise is generated at the start and end of recording is experimentally obtained in advance, and the muting is based on the result. The time can be set accurately and to the minimum necessary. Therefore, unlike the hardware muting circuit, the muting time becomes longer than necessary, and the problem that the voice to be originally recorded is cut does not occur.

Next, another embodiment of the present invention will be described with reference to FIG. FIG. 4 is a conceptual diagram showing a data recording state in a semiconductor memory. Point C is the recording start position of the first recording block, point D is the recording end position of the first recording block and the next second recording block. , The point E indicates the recording end position of the second recording block, and in this embodiment, for example, a recording start marker is provided at the fifth frame from each recording start position C, D, and each recording end position. By recording a recording end marker in the frame 5 frames before D and E, C is recorded in the first recording block.
The data between the S point and the DE point and between the DS point and the EE point is reproduced, and the other data is excluded from the reproduction. Now, assuming that the length of one frame is 20 milliseconds, 100 milliseconds from the beginning of each recording block and 100 milliseconds before the end of each recording block are excluded from reproduction.

Thus, the recording and reproducing operations as described above are
A reproduction control marker recording means for recording a reproduction start marker at an address position after a fixed time after the start of recording and a reproduction end marker at an address position for a predetermined time before the end of recording, and starting reproduction by detecting the reproduction start marker. The reproduction control marker recording means is executed by providing a reproduction control means for ending the reproduction by detecting the end marker.
In the block diagram of, the digital signal processing circuit 6 and the control unit 9 are included, and the reproduction control means is included in the control unit 9.

First, the operation during recording will be described. When the recording switch is turned on, the control unit 9 turns the recording switch O
While judging N, the recording mode is set and the converter 5 and the digital processing circuit 6 are put into the operating state, while the voice is converted into an electric signal (analog signal) by the microphone 1 and is transmitted to the converter 5 via the amplifier 2. After being input and converted into a digital signal, it is input to the digital signal processing circuit 6, and the control unit 9 controls according to the control flowchart of FIG.

In the recording mode state, the control unit 9 first sets the value MF of the number of muting frames to "5", and then sets the frame count number FS to "1". Next, after confirming that the recording switch is ON, the voice data converted into digital signals by the converter 5, that is, the voiced data is compressed, and then written to the semiconductor memory 7 by specifying an address. The number of frame frames MF is compared with the frame count number FS. If the relationship of MF = FS is not satisfied, the frame count number is incremented by "1". After that, such an operation is repeated for each frame until the relationship of MF = FS is satisfied.

Eventually, the frame count number FS becomes "5".
When MF = FS, the reproduction start marker (hereinafter abbreviated as S marker) is recorded at the current memory address of the semiconductor memory 7. The S mark address is stored. After that, the frame count number FS is incremented by “1” and the relationship of MF = FS is lost again, so that the voice data is repeatedly recorded for each frame until the recording switch is turned off.

When the recording switch is turned off,
The control unit 9 determines whether the recording switch is OFF,
S is compared, and if MF ≧ FS, the recording mode state is released.

On the other hand, if the relation of MF ≧ FS is not satisfied, then MF from the current memory address of the semiconductor memory 7, that is,
A reproduction end marker (hereinafter abbreviated as E marker) is recorded by designating the address of 5 frames before, and the recording mode state is released after the E mark address is stored.

Next, the operation during reproduction will be described.
As shown in the control flow chart of FIG.
The marker address is compared with the E marker address, and when S marker address> E marker address, the data of the recording block is not reproduced, and the recording block reproduction operation that is not in the relationship of S marker address> E marker address is performed. The reproduction operation is performed by detecting the S marker.
The voiced data from the marker address data to the next E marker address is reproduced.

Referring to FIG. 4, voiced data between CS and DE and between DS and EE is reproduced, and C and CS are reproduced.
Control is performed so as not to reproduce the voiced data between DE and DS and between EE and E. Occurrence of mechanical noise due to ON / OFF of the recording switch and bias fluctuation noise of the microphone 1, the amplifier 2, etc. is in the period between C and CS, between DE and DS, and between EE and E, and therefore noise occurs. Then, at the time of being recorded in the semiconductor memory 7, those noises are not reproduced at the time of reproduction and no pop noise is output from the speaker 3.

In such an embodiment, S
Since the recording positions of the marker and the E marker can be set finely and accurately according to the time of occurrence of noise, the recorded voiced data is not cut more than necessary.

In the above embodiment, the S marker and the E marker are recorded, but without recording such a marker, for example, reproduction start and end are performed based on the start address and end address of the recording block. The voiced data from the reproduction start address to the end address may be reproduced after obtaining the address.

In addition, the present invention is not limited to the embodiments described above and shown in the drawings, and it goes without saying that the present invention can be appropriately modified and implemented without departing from the scope of the invention.

[0036]

As described above, according to the first aspect of the present invention, the silent noise is forcibly recorded for a certain period of time after the start of recording and before the end of recording. When the bias fluctuation noise of the amplifier or the like is generated, these noises are not recorded in the semiconductor memory, and it is possible to prevent the generation of annoying pop noise at the time of reproduction, and moreover, the muting time is one frame long. Since the minimum time can be set finely and accurately, there is no inconvenience that the muting time becomes longer than necessary and the audio data to be originally recorded is cut like that of a hardware muting circuit, It is possible to provide what is practically excellent in that the generated audio data can be suppressed to the minimum. .

According to the second and third aspects of the invention, the data of each recording block is reproduced from the beginning of the recording for a fixed time and before the end, and mechanical noise and bias are reproduced. When the fluctuating noise is recorded, those noises are not reproduced at the time of reproduction, so that no jarring pop noise is generated.
The playback start and end positions can be set precisely and precisely according to the time when noise occurs, so audio data will not be cut more than necessary, and the cut audio data can be suppressed to a minimum. Can be provided.

[Brief description of drawings]

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

FIG. 2 is a control flowchart for recording in the same device.

FIG. 3 is a conceptual diagram showing an example of a data recording state of a semiconductor memory.

FIG. 4 is a conceptual diagram showing another example of a data recording state of a semiconductor memory.

FIG. 5 is a control flowchart at the time of recording of the audio recording / reproducing apparatus according to another embodiment of the present invention.

FIG. 6 is a control flowchart for reproduction in the same device.

[Explanation of symbols]

 1 Microphone 3 Speaker 5 Converter 6 Digital Signal Processing Circuit 7 Semiconductor Memory 9 Controller

Claims (3)

[Claims] 1. An input audio analog signal is converted into a digital signal, the digital signal is framed at a constant time interval and compressed for each frame unit, and the compressed data is recorded in a semiconductor memory and is reproduced. In a voice recording / reproducing apparatus for decompressing compressed data read from the semiconductor memory for each frame, converting the decompressed digital signal into a voice analog signal, and outputting An audio recording / reproducing apparatus comprising a silent data recording means for recording silent data for a certain period of time before the end. 2. An input audio analog signal is converted into a digital signal, the digital signal is framed at a constant time interval and compressed for each frame unit, and the compressed data is recorded in a semiconductor memory and reproduced. In a voice recording / reproducing apparatus that decompresses the compressed data read from the semiconductor memory for each frame, converts the decompressed digital signal into a voice analog signal, and outputs it, each record recorded in the semiconductor memory during reproduction. An audio recording / reproducing apparatus comprising reproduction control means for reproducing block data from the beginning thereof for a certain period of time and for a certain period of time before the end thereof. 3. The semiconductor memory is provided with a reproduction control marker recording means for recording a reproduction start marker at an address position after a fixed time after the start of recording and a reproduction end marker at an address position before a fixed time before the end of recording. A voice recording / playback apparatus, wherein the playback control means detects a playback start marker to start playback, and ends playback by detecting a playback end marker.