JP2018207318A - Voice processing device, control method therefor, program and storage medium - Google Patents

Abstract

To provide a voice processing device for implementing automatic level control of a voice signal in such a manner that audibility is prevented from being deteriorated even in a case where an excessively low level voice is inputted after a short pronunciation.SOLUTION: A voice processing device comprises: signal processing means for processing an inputted voice signal in such a manner that an amplitude level of the voice signal is settled within a predetermined range between an upper limit threshold and a lower limit threshold; level detection means for detecting the amplitude level of the voice signal that is outputted from the signal processing means; means for controlling a gain value to be set to the signal processing means based on the amplitude level that is detected by the level detection means; and hold means for holding the gain value that is being set to the signal processing means. The control means performs processing for decreasing the gain value to be set in a case where the amplitude level that is detected by the level detection means is changed from a state where the amplitude level is lower than the upper limit threshold to a state where the amplitude level is equal to or higher than the upper limit threshold, and performs processing for increasing the gain value to be set in a case where the amplitude level is changed from a state where the amplitude level is not lower than the lower limit threshold to a state where the amplitude level is equal to or lower than the lower limit threshold.SELECTED DRAWING: Figure 1

Description

  The present invention relates to automatic voice level control.

  2. Description of the Related Art Conventionally, a speech processing apparatus having an automatic level control (ALC) function for controlling the amplitude of an input speech signal to an appropriate level is known. The ALC controls such that the level is suppressed (limit operation) when an excessive level of sound is input than a predetermined threshold, and the level is amplified (recovery operation) when an excessive level of audio is input. I do. In ALC, it is necessary to appropriately control the recovery operation when a small sound (under level sound) is input immediately after a sudden loud sound (over level sound). Patent Document 1 describes a technique for controlling the gain of the level of the audio signal in the recovery operation based on a comparison between the level of the input audio signal and a threshold value.

JP 2004-104692 A

  In the above-mentioned Patent Document 1, when sudden excessive sound (hereinafter referred to as short sound) is continuously input, if an under sound is input immediately after the short sound, the gain of the level of the audio signal is rapidly increased. Recovery operation is performed so as to increase. As a result, the level of the audio signal other than the short sound may fluctuate with a short period, and it may be difficult to hear. In addition, for example, it is difficult to adaptively control the level of the audio signal when the level of the audio signal fluctuates violently such as wind noise.

  The present invention has been made in view of the above problems, and an object of the present invention is to realize automatic level control of an audio signal that does not deteriorate the audibility even when an under-level sound is input after a short sound.

  In order to solve the above problems and achieve the object, the audio processing device of the present invention is set so that the amplitude level of the input audio signal falls within a predetermined range based on the upper limit threshold and the lower limit threshold. Based on signal processing means for processing the audio signal using a gain value, level detection means for detecting the amplitude level of the audio signal output from the signal processing means, and the amplitude level detected by the level detection means A means for controlling a gain value set in the signal processing means, wherein the setting is performed when the amplitude level detected by the level detection means becomes lower than the upper limit threshold or more than the upper limit threshold. A process of reducing the gain value is performed, and the gain value to be set is set when the amplitude level is not lower than the lower limit threshold value or less than the lower limit threshold value. Control means for performing processing, and holding means for holding a gain value set in the signal processing means, wherein the control means is configured such that the amplitude level detected by the level detection means is the upper limit threshold value. When the threshold value is equal to or higher than the upper threshold value from a state other than the above, the holding means is controlled to hold the gain value set before the upper threshold value is exceeded, and the level detection means detects the gain value. After the amplitude level becomes equal to or greater than the upper limit threshold, the state in which the amplitude level is equal to or greater than the upper limit threshold does not continue for a predetermined time or more and the amplitude level is equal to or less than the lower limit threshold. Based on the gain value held by the means, a process of increasing the set gain value is performed with a predetermined correction gain value equal to or lower than the held gain value as an upper limit.

  According to the present invention, it is possible to realize automatic level control of an audio signal that does not deteriorate the audibility even when an under-level sound is input after a short sound.

1 is a diagram illustrating a configuration of a voice processing device according to a first embodiment. 5 is a flowchart showing the operation of the first embodiment. The figure which compares and shows the recovery operation | movement of the conventional and this embodiment when a short pronunciation is input continuously. FIG. 4 is a diagram illustrating a configuration of a sound processing apparatus according to a second embodiment. 9 is a flowchart showing the operation of the second embodiment. The figure which shows the relationship between an air volume and a correction coefficient. The figure which shows the basic composition of a speech processing unit. The figure which compares and shows the recovery operation | movement when the audio | voice signal with a large amplitude level is input for a short time, and when it is input for a long time. The figure which shows the relationship between the gain value set by the process of ALC, and the amplitude level of an audio | voice signal.

  Hereinafter, embodiments of the present invention will be described in detail. The embodiment described below is an example for realizing the present invention, and should be appropriately modified or changed according to the configuration and various conditions of the apparatus to which the present invention is applied. It is not limited to the embodiment. Moreover, you may comprise combining suitably one part of each embodiment mentioned later.

  Hereinafter, as an audio processing apparatus according to this embodiment, an example will be described in which the present invention is applied to an imaging apparatus such as a digital video camera that can capture image data such as still images and moving images and record it together with audio data.

[Recovery operation in automatic level control]
The audio processing apparatus according to the present embodiment automatically adjusts the amplitude level of the audio signal so that the amplitude level indicating the amplitude of the audio signal input from the audio input unit such as a microphone is within a predetermined range that is an appropriate level. It has an automatic level control (ALC) function to adjust in With ALC, it is possible to reduce the overload of a sound output unit such as a speaker to prevent deterioration, and to optimize the dynamic range to improve the sound quality.

  In ALC, signal processing is performed to multiply the audio signal by a gain value so that the amplitude level of the input audio signal falls within a predetermined range that is an appropriate level, and the amplitude level of the audio signal is automatically controlled.

  First, before describing the configuration and functions of the speech processing apparatus according to the present embodiment, the basic configuration of the speech processing apparatus and the recovery operation by ALC for short pronunciation will be described with reference to FIG.

  In FIG. 7, the audio processing device 100 performs a limit operation to decrease the gain value applied to the audio signal when the amplitude level of the input audio signal is larger than the upper limit threshold value X of a predetermined appropriate range. Also, the audio processing device 100 performs a recovery operation to increase the gain value when the amplitude level of the input audio signal is smaller than the lower limit threshold value Y of a predetermined appropriate range.

  The control unit 102 controls operations of a level detection unit 103, a timer unit 104, and a gain setting unit 105, which will be described later. The level detection unit 103 detects the amplitude level of the audio signal output from the signal processing unit 101, and outputs the detection result to the control unit 102. The control unit 102 controls the gain setting unit 105 so as to determine a gain value corresponding to the amplitude level detected by the level detection unit 103. The gain setting unit 105 sets a gain value according to the amplitude level instructed from the control unit 102, and outputs the set gain value to the signal processing unit 101. The signal processing unit 101 multiplies the input audio signal by the gain value set by the gain setting unit 105 to amplify or attenuate the amplitude level.

  In the timer unit 104, the amplitude level of the audio signal detected by the level detection unit 103 suddenly increases (becomes the upper limit threshold X or more), and the time until the level decreases again (the state of the upper limit threshold X or more continues. And the measurement result is output to the control unit 102. The control unit 102 determines whether or not the input sound is a short sound according to the time measured by the timer unit 104.

  The gain setting unit 105 decreases the gain value when the amplitude level of the audio signal amplified by the signal processing unit 101 detected by the level detection unit 103 is larger than the upper limit threshold value X. When the level is larger than the upper limit threshold value X, a plurality of threshold values A, B, and C (A, B, C> X) may be set. Is further reduced. When the amplitude level of the audio signal after being processed by the signal processing unit 101 is smaller than the lower limit threshold Y, the gain value is increased. In the case of a level smaller than the lower threshold Y, a plurality of thresholds D, E, F (D, E, F> Y) may be set, and an audio signal having a larger amplitude level is input. The gain value is increased according to each threshold value (where X> Y).

  FIG. 8 shows a comparison of the recovery operation when a short sound is input (a) and when an excessive level sound is input for a long time (b).

  In the case of FIG. 8A, the control unit 102 determines that the amplitude level detected by the level detection unit 103 is larger than the upper limit threshold value X, and the time of the large amplitude level measured by the timer unit 104 is shorter than the threshold value. It is determined that a short pronunciation has been input. At that time, the timer unit 104 sets the counter value to “ATT_CNT” (ATT_CNT is a numerical value, ATT_CNT> 0) (T1), and periodically decreases (T2), for example. When the amplitude level of the audio signal input before ATT_CNT becomes 0 decreases and the control unit 102 determines that the recovery operation is to be performed, the gain value of the recovery operation is increased to quickly reach an appropriate level. So as to amplify the amplitude level. This is called “first recovery”. Further, after ATT_CNT becomes 0, the increase in the gain value of the recovery operation is reduced (T3 to T4). This is called “slow recovery”.

  In the case of FIG. 8B, the control unit 102 determines that a sound other than the short sound is input because the time of the large amplitude level measured by the timer unit 104 is long. The control unit 102 sets ATT_CNT when the amplitude level changes from a small level to a large level in the timer unit 104 (T1), but does not perform a recovery operation during a period in which a high level sound is input. In the meantime, ATT_CNT becomes 0 at the time (T1 to T3) when a high level sound is input. Thereafter, when the amplitude level of the input audio signal becomes small, ATT_CNT is 0, so that the recovery operation is slow recovery (T4).

  By performing the recovery operation in this way, the input audio signal is controlled to an appropriate level by the signal processing unit 101.

  [Embodiment 1] Next, automatic level control of Embodiment 1 will be described with reference to FIG.

  FIG. 1 shows a configuration of the speech processing apparatus 100 according to the first embodiment. In the following, description of the same configuration as that in FIG. 7 will be omitted, and description will be made focusing on the differences.

  In FIG. 1, a gain value set by the gain setting unit 105 is always input to the gain holding unit 106. As will be described later, the gain holding unit 106 holds the gain value input at that time in response to an instruction from the control unit 102. When it is determined that the input sound is a short sound, the control unit 102 performs control so that the gain setting unit 105 sets the gain value held immediately before the short sound is determined as an upper limit. Therefore, the gain correction unit 107 corrects the gain value set by the gain setting unit 105 so as to be equal to or less than the gain value held by the gain holding unit 106. When the gain correction unit 107 corrects the gain value, the correction coefficient k to be multiplied by the gain value held in the gain holding unit 106 is 1 or less (k ≦ 1). The predetermined correction gain value calculated by the gain correction unit 107 is replaced with the value set in the gain setting unit 105, and the signal processing unit 101 multiplies the audio signal by the gain value and outputs the result.

  <Automatic Level Control Flow> Next, the operation of the speech processing apparatus 100 of this embodiment will be described with reference to FIG. 2 is realized when the control unit 102 reads out and executes a control program stored in a memory (not shown) and controls each unit of the audio processing apparatus 100. The same applies to FIG. 5 described later.

  In the following, it is assumed that the audio processing device 100 performs a process of controlling the level of the input audio signal using the gain value and outputs the audio signal.

  In S <b> 201, the control unit 102 detects the amplitude level of the audio signal processed by the signal processing unit 101 based on the gain value set by the gain setting unit 105 by the level detection unit 103.

  In S202, the control unit 102 determines whether or not the amplitude level of the audio signal processed by the signal processing unit 101 detected in S201 is equal to or higher than an upper limit threshold within a predetermined appropriate range. If the control unit 102 determines that the amplitude level of the audio signal is greater than or equal to the upper limit threshold value, the control unit 102 proceeds to S203, and if it determines that the audio signal amplitude level is less than the upper limit threshold value, ends the process.

  In step S <b> 203, the control unit 102 instructs the gain holding unit 106 to hold the gain value set by the gain setting unit 105. In addition, when a short sound is input, the control unit 102 controls the gain holding unit 106 so as to hold the gain value immediately before the short sound. The holding time is until the next short pronunciation is input.

  In S <b> 204, the control unit 102 causes the timer unit 104 to measure the time during which the state in which the amplitude level of the audio signal is equal to or higher than the threshold value continues for the threshold value or higher.

  In step S <b> 205, the control unit 102 determines whether the input sound is a short pronunciation based on the time measured by the timer unit 104. If the time during which the state where the amplitude level of the audio signal is equal to or greater than the threshold is less than the predetermined time, the control unit 102 determines that the sound is short and proceeds to S206. If it continues for more than the time, it is determined that the sound is not short and the process proceeds to S207.

  In S206, the control unit 102 corrects the gain value by multiplying the gain value held in S203 by the correction coefficient k (k ≦ 1) in the gain correction unit 107.

  In S207, if the control unit 102 determines that a short sound is input in S205, the gain setting unit 105 sets the gain value corrected by the gain correction unit 107 in S206 as the upper limit value of the gain during the recovery operation. Set to. If the control unit 102 determines in S205 that the sound is not short, the gain setting unit 105 does not set the corrected gain value as the upper limit value during the recovery operation, as will be described later with reference to FIG. Set up.

  Here, with reference to FIG. 9, the process of setting the gain value of ALC when the correction gain value is not set as the upper limit value of the gain during the recovery operation will be described. FIG. 9 shows the relationship between the gain value set by the gain setting unit 105 and the amplitude level of the audio signal detected by the level detection unit 103.

  The gain setting unit 105 decreases the gain value when the amplitude level of the audio signal detected by the level detection unit 103 is greater than the upper threshold X, and decreases the gain value as the amplitude level of the audio signal increases. . Specifically, it is set to a value between the normal value Gn and a smaller Glim (Gn> Glim). Further, when the amplitude level of the audio signal is smaller than the lower limit threshold Y, the gain value is increased, and the gain value is increased as the amplitude level of the audio signal is decreased (X> Y). Specifically, the gain value is set to a value between the normal value Gn and a larger Grec (Grec> Gn). However, the gain value has Grec as an upper limit value and Glim as a lower limit value. When the amplitude level of the audio signal is other than that, the gain value is fixed to the normal value Gn.

  Returning to FIG. 2, in S <b> 208, the control unit 102 performs a recovery operation using the gain value set by the gain setting unit 105.

  The recovery operation is a process in which the signal processing unit 101 amplifies the amplitude level of the audio signal based on the gain value set by the gain setting unit 105. For example, when the amplitude level of the audio signal is smaller than the lower limit threshold Y in FIG. 9, the control unit 102 increases the gain value by the gain setting unit 105 and amplifies the amplitude level by the signal processing unit 101.

  FIG. 3 shows a comparison between the recovery operation of the present embodiment and the conventional case where short pronunciations are continuously input. 3A shows the waveform change of the input sound, FIG. 3B shows the waveform change of the output sound due to the conventional recovery operation, and FIG. 3C shows the change of the gain value in the conventional recovery operation. It is assumed that the input sound in FIG. 3A is input alternately or randomly when the amplitude level of the sound between consecutive short sounds is small and large.

  When a short pronunciation as shown in FIG. 3A is input, the level detection unit 103 determines that the amplitude level of the audio signal from the signal processing unit 101 is larger than the upper limit threshold. A limit operation is performed to control the gain so that it is below the upper threshold. When the level detection unit 103 determines that the amplitude level of the audio signal from the signal processing unit 101 is greater than the upper limit threshold, the control unit 102 instructs the gain setting unit 105 to decrease the gain value. put out. When the amplitude level of the audio signal from the signal processing unit 101 becomes lower than the upper limit threshold, the gain setting unit 105 is controlled to hold the gain.

  Further, when the amplitude level of the audio signal from the signal processing unit 101 becomes smaller than the lower limit threshold, the control unit 102 controls the gain setting unit 105 so as to gradually increase the gain. For example, as shown in FIG. 9, the gain value is set not to be larger than the gain value Grec. As the amplitude level of the audio signal after the limit operation is smaller, the gain value is increased more rapidly. As the amplitude level of the audio signal after the limit operation is increased, the gain increase is made slower. However, it is assumed that the amplitude level of the audio signal during the recovery operation does not exceed the upper threshold.

  When an under-level sound is input immediately after a short sound as shown in FIG. 3 (a), a recovery operation is performed to quickly set the amplitude level of the input sound to an appropriate level, and as shown in FIG. 3 (c), It is necessary to increase the gain value. Further, when a sound having a large amplitude level (a sound whose amplitude level is smaller than that of the short sound and equal to a level within a predetermined appropriate range) is input immediately after the short sound, the gain value is decreased. In recovery operation, if an under-level sound is input immediately after a short sound, if the gain value is increased rapidly, a suddenly loud sound may be heard or the sound may be unstable. Sense of hearing tends to deteriorate due to difficulty and discomfort.

  FIG. 3 (d) shows the waveform change of the output sound due to the recovery operation when an under-level sound is input immediately after the short sound in FIG. 3 (a), and FIG. 3 (e) shows the gain in the recovery operation of this embodiment. Each change in value is shown. As shown in FIG. 3E, the sound processing apparatus 100 according to the present embodiment holds the gain value before the short sound is input in the gain holding unit 106. Then, in the recovery operation when an under-level sound is input immediately after a short sound, the gain value set by the gain setting unit 105 is controlled with the gain value held by the gain holding unit 106 as the upper limit.

  For example, in the case of short pronunciation, the amplitude level of the audio signal increases before ATT_CNT in FIG. Therefore, the fast recovery operation is performed. At this time, the control unit 102 uses the correction gain value obtained by correcting the gain value held in the gain holding unit 106 by the gain correction unit 107 as the upper limit value of the gain during the recovery operation. The gain setting unit 105 is set. During the fast recovery operation, the gain is gradually increased by the gain setting unit 105, and when the gain reaches the upper limit value corrected by the gain correction unit 107, the gain value is not increased thereafter. Therefore, during the recovery operation, the gain value is set in the gain setting unit 105 with the gain value held in the gain holding unit 106 as the upper limit.

  In this way, by setting an upper limit on the gain value for the recovery operation, when an under-level sound is input after a short sound, the previous gain value is retained, and the gain value retained in the recovery operation is retained. Control is performed so that the recovery operation is performed at a higher level. This can be expected to improve the audibility of the output sound.

  [Embodiment 2] Next, automatic level control of Embodiment 2 will be described with reference to FIGS.

  In the first embodiment, a short pronunciation is input to the input sound, and an under-level sound is input after the short pronunciation. However, in the second embodiment, wind noise (hereinafter, wind noise) is input after the short sound is input. The case will be described.

  FIG. 4 shows the configuration of the speech processing apparatus 100 of the second embodiment. FIG. 5 is a flowchart illustrating the operation of the speech processing apparatus 100 according to the second embodiment. In the following description, the same components as those in FIG. 1 and the processes similar to those in FIG. 2 are denoted by the same reference numerals, description thereof will be omitted, and differences will be mainly described.

  When level control by ALC is performed on the input voice, if wind noise is input, it also affects the voice and music input together with the wind noise. When the wind noise is large, the gain value is reduced by the ALC limit operation. However, since the voice and music input together with the wind noise are also subjected to the limit operation, the sound becomes small. Also, if the wind suddenly stops, only the voice and music will be, so if the amplitude level of the voice or music is lower than the threshold, it will be fast recovery operation, increase the gain value for the voice and music, Increase the amplitude level. In this way, when the amplitude level of the input sound fluctuates greatly due to wind noise being input or not being input, the amplitude level of the voice or music is not constant, and it may be heard as an unstable sound, and listening is not possible. It may be difficult or uncomfortable, and hearing may worsen.

  Therefore, in the present embodiment, the sound processing device 100 is provided with the air volume detection unit 401 and the correction value adjustment unit 402 so that the gain value is corrected in consideration of a case where wind noise is input after a short sound. Yes.

  In FIG. 4, the air volume detection unit 401 detects the level of wind noise included in the input sound with a filter circuit or the like. For example, when left and right two-channel audio signals from a stereo microphone are input to the audio processing device 100, the air volume is detected using the difference between the right and left channel audio signals. Specifically, the air volume detection unit 401 outputs, as air volume information, a signal indicating that the air volume is large (the wind noise is large or large) when the absolute value of the difference signal between the left and right channels is greater than a predetermined threshold. Further, the air volume detection unit 401 outputs a signal indicating that the air volume is low (wind noise is low or low) as the air volume information when the absolute value of the difference signal is smaller than a predetermined threshold. As described above, when there is wind noise, the amount of wind noise is detected by using the characteristic that the phase difference between the two audio signals increases in the low frequency component (for example, 100 Hz or less).

  The correction value adjustment unit 402 adaptively changes the value (correction value) of the correction coefficient k of the gain correction unit 107 according to the air volume information detected by the air volume detection unit 401. FIG. 6 shows the relationship between the air volume and the correction coefficient k of the gain value. When the air volume detection unit 401 detects that a large amount of air is input, the gain coefficient correction coefficient k is adjusted to a small value, and when the air volume is input to a small amount, the correction coefficient k is adjusted to a large value.

  Next, the operation of the speech processing apparatus 100 of this embodiment will be described with reference to FIG.

  In S501, the control unit 102 determines whether or not wind noise is detected by the air volume detection unit 401. If wind noise is detected, the process proceeds to S502, and if wind noise is not detected, the process proceeds to S503.

  In S <b> 502, when wind noise is detected, the control unit 102 adaptively changes the value of the correction coefficient k of the gain correction unit 107 by the correction value adjustment unit 402 according to the detected amount of wind noise.

  Thereafter, in S503 to S510, processing similar to S201 to S208 in FIG. 2 is performed.

  According to the present embodiment, as the air volume increases, the maximum gain value can be reduced as shown in FIG. 6, so that the amplitude level of the processed main voice (voice or music) becomes unstable due to wind noise. It can be suppressed.

  In this embodiment, the air volume is detected based on the difference signal between the two audio signals. However, the present invention is not limited to this, and the low frequency level of the two audio signals is simply compared with a predetermined air volume threshold. You may ask. Alternatively, the air volume may be detected by using a device capable of detecting air displacement such as a wind pressure sensor, an anemometer, or a device using a laser and a camera.

  In addition, the change in the gain correction coefficient according to the value and the air volume shown in FIG. 6 is not limited to these, and can be tuned in a scene where ALC is performed and set to an appropriate value. To do.

  In addition, the audio processing device of each embodiment can be applied to an imaging device such as a digital video camera, but is not limited thereto. The sound processing apparatus according to each embodiment can be applied to any apparatus as long as it records or inputs and outputs external sound. For example, it may be a mobile phone or a type of smartphone (including eyeglass-type terminals and wristwatch-type terminals), a tablet terminal, a voice recorder, a media player, and the like.

[Other Embodiments]
The present invention supplies a program that realizes one or more functions of the above-described embodiments to a system or apparatus via a network or a storage medium, and one or more processors in a computer of the system or apparatus read and execute the program This process can be realized. It can also be realized by a circuit (for example, ASIC) that realizes one or more functions.

DESCRIPTION OF SYMBOLS 100 ... Voice processing apparatus 101 ... Signal processing part 102 ... Control part 103 ... Level detection part 104 ... Timer part 105 ... Gain setting part 106 ... Gain holding part 107 ... Gain correction part 401 ... Air volume detection , 402... Correction value adjustment unit

Claims (9)

Signal processing means for processing the audio signal using a set gain value so that the amplitude level of the input audio signal falls within a predetermined range by an upper limit threshold and a lower limit threshold;
Level detection means for detecting the amplitude level of the audio signal output from the signal processing means;
A means for controlling a gain value set in the signal processing means based on the amplitude level detected by the level detection means, wherein the amplitude level detected by the level detection means is lower than the upper limit threshold value. The gain to be set when the gain level to be set is reduced when the threshold value is equal to or higher than the upper threshold value, and when the amplitude level is not lower than the lower threshold value or lower than the lower threshold value Control means for performing processing to increase the value;
Holding means for holding a gain value set in the signal processing means,
The control means, when the amplitude level detected by the level detection means is not equal to or higher than the upper limit threshold, when the amplitude level is equal to or higher than the upper limit threshold, the gain value set before the upper limit threshold or higher is set. Controlling the holding means to hold
After the amplitude level detected by the level detection means becomes equal to or higher than the upper limit threshold, the amplitude level is equal to or lower than the lower limit threshold without continuing the state of being equal to or higher than the upper limit threshold for a predetermined time or longer. In this case, a process of increasing the set gain value with an upper limit of a predetermined correction gain value equal to or less than the held gain value based on the gain value held by the holding means is performed. Voice processing device.   After the amplitude level detected by the level detection unit becomes equal to or higher than the upper limit threshold, and after the state where the amplitude level is equal to or higher than the upper limit threshold continues for a predetermined time or longer, the control means The audio processing apparatus according to claim 1, wherein when the threshold value is equal to or lower than the lower limit threshold value, the process of increasing the set gain value is performed without setting the correction gain value as an upper limit. A timer means;
The audio processing apparatus according to claim 1, wherein the control unit measures a time during which the amplitude level detected by the level detection unit is equal to or greater than the upper limit threshold by the timer unit.   The audio processing apparatus according to claim 1, wherein the control unit determines the correction gain value by multiplying a gain value held in the holding unit by a predetermined correction value of 1 or less. An air volume detecting means for detecting an amount of wind noise included in the input audio signal;
The audio processing apparatus according to claim 1, wherein the control unit further determines the correction gain value based on an amount of wind noise detected by the air volume detection unit.   The control means determines the correction gain value by multiplying the gain value held in the holding means by a predetermined correction value of 1 or less, and the amount of wind noise detected by the air volume detection means is large. The speech processing apparatus according to claim 5, wherein the predetermined correction value is set to a smaller value. Signal processing means for processing the audio signal using a set gain value so that the amplitude level of the input audio signal falls within a predetermined range by an upper limit threshold and a lower limit threshold;
Level detection means for detecting the amplitude level of the audio signal output from the signal processing means;
Control means for controlling a gain value set in the signal processing means based on the amplitude level detected by the level detection means;
Holding means for holding a gain value set in the signal processing means, and a control method for a sound processing apparatus comprising:
When the amplitude level detected by the level detection means becomes lower than the upper limit threshold from a state lower than the upper limit threshold, the gain value to be set is reduced, and the amplitude level is lower limit threshold A control step for performing a process of increasing the gain value to be set when the lower limit threshold or less is reached from a state not lower than
In the control step, when the amplitude level detected by the level detection unit is not equal to or higher than the upper limit threshold and is equal to or higher than the upper limit threshold, the gain value set before the upper limit threshold or higher is set. Controlling the holding means to hold
After the amplitude level detected by the level detection means becomes equal to or higher than the upper limit threshold, the amplitude level is equal to or lower than the lower limit threshold without continuing the state of being equal to or higher than the upper limit threshold for a predetermined time or longer. In this case, a process of increasing the set gain value with an upper limit of a predetermined correction gain value equal to or less than the held gain value based on the gain value held by the holding means is performed. Control method for a voice processing apparatus.   A program for causing a computer to execute the control method according to claim 7.   A computer-readable storage medium storing a program for causing a computer to execute the control method according to claim 7.

Images