JPH07168599A - Voice recognizing and reproducing device - Google Patents

Abstract

PURPOSE:To suppress degradation in voice quality of a reproduced voice influ enced by ambient environment to the minimum. CONSTITUTION:This device is provided with a frame energy calculating section 4 which calculates frame energy E of a digital voice signal inputted dividing into frame intervals previously decided, a sound/silence discriminating section 5 which discriminates with a threshold value whether an output signal from the frame energy calculating section 4 is sound or silence, a system control section 12 in which frames discriminated as sound are encoded to ADPCM, but silence frames are not encoded, and which counts the number of silence frames continued, and a reproduction volume which can adjust variably and arbitrarily the threshold value of the silence discriminating section 5 monitoring a reproduced sound.

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.

[0002]

2. Description of the Related Art Conventionally, a voice recording / reproducing apparatus called a so-called tape recorder using a magnetic tape as a recording medium has been widely used. As described above, the device using the magnetic tape has an advantage of being able to record a voice for a long time, but has a drawback that it is difficult to downsize the device and it takes time to access the voice.

In order to solve this drawback, in recent years, an audio recording / reproducing apparatus called a so-called digital recorder using a semiconductor memory as a recording medium is being developed. Since the digital recorder does not require a mechanism, it can be easily miniaturized, and since the recording medium is housed in a removable audio cartridge, it has many handling advantages.

By the way, in such a digital recorder, the voice data is encoded and recorded in the semiconductor memory, but in order to store the voice data in the memory more efficiently, the silent section is not encoded. One in which only the voiced section is encoded and only the duration information is recorded, and one in which the encoding bit allocation amount is switched between the voiced section and the silent section have been devised. Japanese Patent Laid-Open No. 4-16
It is disclosed in the 1999 publication.

Further, as a method that has been often used as a method for detecting a voiced / non-voiced area, there is known a method that pays attention to a voice power or an autocorrelation function and judges the threshold value. Furthermore, Proceedings of the Acoustical Society of Japan 3-7-15
(Takizawa, Norimatsu et al., Published in March, 1991) discloses a method for automatically adjusting a threshold value in accordance with a background noise level.

[0006]

However, such a method using the voice power and the autocorrelation function has a high dependency on the speaker and the surrounding environment, and it is difficult to detect a stable voiced / silent section. However, there is a problem that the sound quality of the reproduced voice is greatly deteriorated due to the erroneous detection.

Further, even in the method of automatically adjusting the threshold value in accordance with the background noise level, there is a problem in that it is not possible to automatically follow the actual differences in various surrounding environments and it is easy to make an erroneous detection. It was

The audio recording / reproducing apparatus of the present invention has been made in view of such a problem, and an object thereof is to minimize the deterioration of the sound quality of reproduced audio due to the influence of the surrounding environment. An object is to provide an audio recording / reproducing device capable of performing the same.

[0009]

In order to achieve the above-mentioned object, according to the present invention, an input audio signal is converted into a digital signal, encoded, recorded in a semiconductor memory, and recorded in the semiconductor memory. In a voice recording / reproducing apparatus that decodes a signal, converts it into an analog signal, and reproduces and outputs it, characteristic parameter calculation for obtaining characteristic parameters of a digital signal input by dividing it into predetermined frame intervals Means, a sound / silence judging means for threshold value judgment of whether the output signal from the characteristic parameter calculating means is related to sound or silence, and a judgment result by the sound / silence judging means. The coding bit distribution amount switching means for switching the coding bit distribution amount based on the above, and the threshold value adjusting means capable of variably adjusting the threshold value of the voice / silence judging means.

The threshold value adjusting means automatically changes the threshold value by a predetermined operation for a predetermined time.
A trial recording / reproducing means for repeating the operation of reproducing the contents after performing the trial recording and a threshold value selecting means for stopping the above operation by an external operation and setting a threshold value at that time are provided.

[0011]

That is, in the audio recording / reproducing apparatus of the present invention, the characteristic parameters of the digital audio signal input by dividing the audio recording / reproducing apparatus into predetermined frame intervals are obtained, and the characteristic parameters are converted into the sound. Whether it is related or silence is determined by a threshold value, and the coded bit allocation amount is switched based on the determination result. The threshold value at this time can be variably adjusted.

Further, when the threshold value is adjusted, the operation of reproducing the contents after performing a trial recording for a predetermined time while changing the threshold value automatically by a predetermined operation is repeated, and the above operation is performed by an external operation. Stop and set the threshold at that time.

[0013]

Embodiments of the present invention will now be described in detail with reference to the drawings. FIG. 1 is a diagram showing the configuration of an audio recording / reproducing apparatus according to the first embodiment. In the figure, the microphone 1 is connected to the frame energy calculation unit 4 via a microphone amplifier 2 and an A / D converter 3. The frame energy calculation unit 4 is connected to the first terminal of the encoding / decoding unit 9 via the voice / silence determining unit 5.

The second terminal of the encoder / decoder 9 is a D / A
The third terminal is connected to the input terminal of the converter 8, and the third terminal is connected to the audio memory unit 11 via the memory control unit 10. The output terminal of the D / A converter 8 is connected via a resistor R1 to one end of a resistor R2 whose other end is grounded and one end of a switch S3.

The output terminal of the D / A converter 8 is further connected to the b-side contact of the switch S1, and this a-side contact is connected to the power source + V (for example, 5 volts). Further, the fulcrum of this switch S1 is grounded at the other end, and the variable resistor VR is connected.
Is connected to one end of a playback volume.

Further, the speaker 6 is connected to a power amplifier 7, and the power amplifier 7 is connected to the other end of the switch S3 and the c-side contact of the switch S2. Further, the d-side contact of the switch S2 is the system control unit 1
2 and its fulcrum is connected to the variable terminal of the reproduction volume.

Here, the switches S1 to S3 and the resistors R1 and R2 constitute a monitor means. Further, the system control unit 12 is connected to the operation input unit 13 and also has switches 1, 2 and 3 for controlling each unit.
Switch 3, voice / silence determination unit 5, and encoding / decoding unit 9
Is connected to the memory control unit 10 and the voice memory unit 11.

In the above structure, the microphone 1 is used during recording.
The voice supplied from the device is amplified by the microphone amplifier 2 and then converted into digital data S (i) by the A / D converter 3 at a sampling frequency of, for example, 8 kHz. This digital data S (i) is divided into a predetermined frame interval (for example, 20 ms is used, which corresponds to 160 samples) and is expressed by the frame energy calculating section 4 by the following equation. Frame energy E is required.

[0019]

[Equation 1]

Where N is the frame length (ie 160
(Sample) is shown. The obtained frame energy E is compared with a predetermined threshold value in the voice / silence determining unit 5, and if E is larger than the threshold value, the frame voice is voiced, otherwise the voice is silent. To determine. Here, the threshold value can be changed by the operation of the user to correspond to the recording environment, which will be described later. Voice /
The encoding / decoding unit 9 performs a predetermined encoding on the input voice signal for which silence has been determined. For example, for a frame that is determined to be voiced, ADPCM (Adaptive Differential Pulse C
ode Modulation) is encoded, but silent frames are not coded and the number of continuous silent frames is counted. The encoded data is recorded in the audio memory unit 11 via the memory control unit 10.

On the other hand, at the time of reproduction, of the recorded data, the data to be reproduced is read from the audio memory section 11 via the memory control section 10 and decoded by the encoding / decoding section 9. After that, the D / A converter 8 restores the analog voice signal. At this time, the switch S1 is held in the b side, the switch S2 is held in the c side, and the switch S3 is held in the open state by the control signal from the control unit 12. Therefore, in this case, the analog audio signal passes through the reproduction volume constituted by the variable resistor VR and then the power amplifier 7
After being amplified by, it is output as sound from the speaker 6.

The operation timings of the above-mentioned recording and reproduction, and stop, etc. are controlled by the system control unit 12 in response to the signal sent from the operation input unit 13. The threshold value adjusting operation according to the first embodiment will be described below with reference to the flowchart of FIG.

First, when the user performs a recording operation from the operation input section 13 (step S1), the control operation of the system control section 12 closes the switch S1 to the a side and the switch S2 to the d side in FIG. After the switch S3 is closed (step S2), the recording operation is started.

As a result, the encoded data obtained by the encoding / decoding unit 9 is recorded in the audio memory unit 11 via the memory control unit 10 as described above.
Since the switch S3 is closed by the operation of, the coded data recorded in the voice memory section 11 at the same time is decoded by the coding / decoding section 9 and returned to the analog voice signal by the D / A converter 8. After that, it is amplified by the power amplifier 7 via the resistor R1 and is output as sound from the speaker 6.
As a result, the user can immediately listen to and monitor the sound obtained by decoding the recorded sound (step S).
3). Here, it is also possible to use an earphone or the like instead of the speaker 6 to monitor the voice.

Next, in step S4, the monitor sound is listened to and it is judged whether or not the sound quality of the monitor sound is good, that is, it is judged whether or not the sound quality is deteriorated due to the adverse effect of background noise. Adjust the volume (step S5). At this time, since the switch S1 is closed to the side a and the switch S2 is closed to the side d, the system controller 12 is supplied with a voltage value according to the position of the reproduction volume at that time. The system control unit 12 converts this voltage value into a digital value, and, for example, refers to a table as shown in FIG. 5 and sets a threshold value corresponding to the voltage value obtained by the adjustment to the sound / silence determination unit 5. Reset (step S6). Then, the process returns to step S4 again, and is repeated until the sound quality of the monitor sound becomes the best.

The second embodiment of the present invention will be described below. In the first embodiment described above, the threshold value was adjusted during the recording operation, but in this embodiment, the trial recording is performed while changing the threshold value for a predetermined time before recording, and according to the result. The threshold value is determined. FIG. 3 is a diagram showing the configuration of the second embodiment, in which the switches S1 to S3 and the resistors R1 and R2 constituting the monitor means shown in FIG. 1 are omitted.

The threshold value adjusting operation according to the second embodiment will be described below with reference to the flowchart of FIG. First, when the user performs a trial recording operation (for example, simultaneously pressing a recording button (not shown) and a threshold value adjusting button arranged on the operation input unit 13) (step ST1), a count value t of a counter (not shown) is set. Is set to 1 and the count value i of the other counter is set to 0 (step ST2). Here, for example, a threshold value (26 dB in this case) corresponding to the count value t = 1 as shown in the table of FIG. 6 is set in the sound / silence determination unit 5, and a predetermined time (for example, 3 seconds) is set in step ST4. Make a trial recording.

In the table shown in FIG. 6, ten threshold values are prepared, so the count value t is from 1 to 1.
It can be changed to 0. Next, step ST5
After incrementing the count value i by 1 in step ST6, step ST6
It is determined whether or not the count value i is 10 or less. If it is 10 or less, the process proceeds to step ST7, increments the count value t by 1 and proceeds to step ST3.

On the other hand, in step ST6, the count value i is 1
If it exceeds 0, the process proceeds to step ST8 and the count value i
Is 20 or less. If it is 20 or less, the count value t is decremented by 1 in step ST9, the process proceeds to step ST3, and the above steps are repeated until the count value i reaches 20. By the above operation, the threshold value in 10 steps is changed at intervals of 3 seconds, and each threshold value is recorded twice.

Next, when the count value i exceeds 20, the process proceeds to step ST10 to set the count values to t = 1 and i = 0, and at the next step ST11 to set the threshold value corresponding to the count value t. To do. Next, the content of the trial recording is reproduced for, for example, 3 seconds (step ST12).
In the next step ST13, the user listens to the reproduced content and determines whether or not the reproduced sound quality is good. If the sound quality is not good, the count value i is incremented by 1 (step ST14), and it is determined in the next step ST15 whether the count value i is 10 or less. If it is 10 or less, the count value t is incremented by 1 in the next step ST16 and the process proceeds to step ST11.

On the other hand, when the count value i exceeds 10 in step ST15, the process proceeds to step ST17, and the count value t is counted in step ST18 until the count value i reaches 20.
Is decreased by 1, and the process proceeds to step ST11.

When the user determines in step ST13 that the sound quality is good, the process proceeds to step ST19, where the threshold value adjustment button is pressed to hold the threshold value corresponding to the count value t at that time.

When the count value i exceeds 20 in step ST17, the process proceeds to step ST20.
For example, a threshold value corresponding to the count value t = 5 is automatically set in the sound / silence determination unit 5.

Incidentally, in the above-mentioned second embodiment, a button which is not used except during the recording operation, for example, an I mark (Instruction mark) button for recording a cue signal is pressed before recording. Therefore, it is possible to allow the above threshold adjustment for a predetermined time (for example, 10 seconds).

As described above, the audio recording / reproducing apparatus according to the first embodiment converts the input audio signal into a digital signal, encodes it, records it in the semiconductor memory (audio memory unit 11), and stores it in the semiconductor memory. In a voice recording / reproducing apparatus that decodes a recorded signal, converts it to an analog signal, and reproduces and outputs it, a characteristic parameter for obtaining a characteristic parameter of a digital signal divided into predetermined frame intervals. -The means for calculating the frame energy, ie, the frame energy calculating section 4 for calculating the frame energy E of the inputted digital signal, and the output signal from the frame energy calculating section 4 as the means for calculating the characteristic parameters is related to the presence of voice or silence. Based on the judgment result by the sound / silence judgment unit (sound / silence judgment unit 5) and the sound / silence judgment unit. Coded bit allocation amount switching means for switching the encoded bit allocation amount, that is, system control for counting the number of continuous silent frames without ADPCM coding for frames determined to be voiced The unit 12 and threshold value adjusting means (reproduction volume and system control section 12) capable of variably adjusting the threshold value of the sound / silence judging means are provided.

Further, in the configuration of the first embodiment described above, the threshold value adjusting means is a monitor means (switches S1 to S3 and a resistor R for monitoring the decoded voice signal).
1, R2).

Further, in the structure of the first embodiment described above, the threshold value adjusting means includes a reproducing volume adjusting means (reproducing volume). Further, in the configuration of the first embodiment described above, the threshold adjusting means according to the second embodiment is
An operation of performing a trial recording for a predetermined time (for example, 3 seconds) while automatically changing the threshold value by pressing the record button and the threshold value adjusting button at the same time by a predetermined operation, and then reproducing the content. Repeating recording / reproducing means (system control section 12) and threshold selecting means for setting the threshold value at that time by stopping the above operation by external operation, for example, pressing down a threshold value adjusting button ( System control unit 12).

Furthermore, in the configuration of the first embodiment described above, the control means (system control unit 12) according to the second embodiment.
Is a condition that a predetermined operation button is operated, that is, when a button that is not used except during the recording operation, such as an I mark button, is pressed,
The threshold value is adjusted by the reproduction volume adjusting means for a predetermined time (for example, 10 seconds).

According to the voice recording / reproducing apparatus in the first embodiment described above, the user can set the threshold value for voiced / non-voiced determination while monitoring the sound quality of the reproduced voice. The quality of the reproduced voice can be minimized.

Further, according to the audio recording / reproducing apparatus in the second embodiment described above, the threshold value can be adjusted before the actual recording, so that the reproduced audio from the beginning of the recorded data is influenced by the surrounding environment. Sound quality deterioration can be minimized.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration of an audio recording / reproducing apparatus according to a first embodiment.

FIG. 2 is a flowchart showing a threshold value adjusting operation according to the first embodiment.

FIG. 3 is a diagram showing a configuration of an audio recording / reproducing apparatus according to a first embodiment.

FIG. 4 is a flowchart showing a threshold adjustment operation according to the second embodiment.

FIG. 5 is a table showing a voltage value of a reproduction volume and a threshold value set according to the voltage value.

FIG. 6 is a table showing a counter value t and a threshold value set accordingly.

[Explanation of symbols]

1 ... Microphone, 2 ... Microphone amplifier, 3 ... A / D converter, 4
... Frame energy calculation unit, 5 ... Voice / silence determination unit,
6 ... speaker, 7 ... power amplifier, 8 ... D / A converter,
9 ... Encoding / decoding section, 10 ... Memory control section, 11 ... Voice memory section, 12 ... System control section, 13 ... Operation input section.

Claims (5)

[Claims] 1. A voice recording in which an input voice signal is converted into a digital signal, encoded and recorded in a semiconductor memory, the signal recorded in the semiconductor memory is decoded, converted into an analog signal and reproduced and output. In the reproducing apparatus, characteristic parameter calculating means for obtaining characteristic parameters of a digital signal divided into predetermined frame intervals and inputted, and an output signal from the characteristic parameter calculating means is sounded. Sound / silence determination means for threshold value determination as to whether it is related to sound or silence, and coded bit distribution amount switching means for switching the coded bit distribution amount based on the determination result by this sound / silence determination means. And a threshold value adjusting means capable of variably adjusting the threshold value of the sound / silence judging means. 2. The audio recording / reproducing apparatus according to claim 1, wherein the threshold value adjusting means comprises a monitor means for monitoring the decoded audio signal. 3. The audio recording / reproducing apparatus according to claim 1, wherein the threshold value adjusting means includes a reproducing volume adjusting means. 4. The threshold value adjusting means includes a trial recording / reproducing means for repeating the operation of reproducing the contents after performing a trial recording for a predetermined time while automatically changing the threshold value by a predetermined operation. 3. The audio recording / reproducing apparatus according to claim 2, further comprising a threshold value selecting means for stopping the operation by operation and setting a threshold value at that time. 5. A control means for enabling threshold adjustment by a reproduction volume adjusting means for a predetermined time before recording on condition that a predetermined operation button is operated. The audio recording and reproducing apparatus described.