JP4604836B2 - Voice processing device, communication device, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a speech processor capable of performing signal processing to a voice signal by following the change of the voice, even if the voice whose volume level often changes. <P>SOLUTION: The speech processor equipped in a multi-function product measures the signal level Lc of the voice signal which is transferred between the multi-function product and communication destination equipment (S145), and updates a sound parameter group (S165) when the absolute value (¾Lp-Lc¾) of difference between a peak value Lp stored in an RAM and the signal level Lc measured in the processing of S145 is larger than a predetermined threshold Lh (S150:yes) and also the signal level Lc measured in the processing of S145 is larger than -30dBm (S155; yes). As a result, the optimizing of the sound volume and the cutting of noise component corresponding to the volume and the canceling of echo, etc., are applied to the voice signal according to the sound volume, and the voice signal which is transferred between the product and the communication destination equipment is corrected to the voice easy to hear. <P>COPYRIGHT: (C)2007,JPO&amp;INPIT

Description

  The present invention relates to a voice processing device that performs signal processing for changing a volume, a sound quality, and the like on a voice signal transmitted to a communication destination device, a communication device including the voice processing device, and The present invention relates to a program for causing a computer to function as an audio processing device.

2. Description of the Related Art Conventionally, for example, a device described in Patent Document 1 below is known as a device for changing a volume for audio transmitted to and from a communication destination device.
The device described in Patent Document 1 determines whether or not the sound is based on the high regressibility of the sound data, and in the case of the sound, a plurality of devices included in a period of about several hundred milliseconds are included. The frame energy is integrated for the frame, and the volume level is determined based on the integrated value.
JP-A-7-177085

  However, in the technique described in Patent Document 1, a change in volume level cannot be detected unless a period of about several hundred milliseconds has elapsed, and the timing at which the volume level actually changes and the change are detected. There was a difference of at least several hundred milliseconds between the detection and the timing of changing the volume. For this reason, for example, when the volume often changes, such as an inflection sound, the volume change control cannot follow the actual volume change well, and the volume may not always be easy to hear.

  Further, in the case of the technique described in Patent Document 1, even if the sound is extremely low, if the sound is determined to be sound, control of changing the sound volume is performed. In some cases, the sound volume may be changed even to a very low volume sound that may not normally be heard.

The present invention has been made to solve the above problems, its object is directed to a voice with a certain level of volume, performing signal processing on the audio signal following the change in the volume An object of the present invention is to provide a voice processing apparatus capable of It is another object of the present invention to provide a communication device including the above-described voice processing device and a program for causing a computer to function as the above-described voice processing device.

The characteristic configuration employed in the present invention will be described below.
The audio processing apparatus of the present invention includes a measuring unit that measures a volume level of an audio signal transmitted from a communication destination device, and a peak value storage unit that can store a peak value of the volume level measured by the measuring unit. First determination means for determining to update the peak value stored in the peak value storage means when the peak value of the volume level measured by the measurement means is greater than a predetermined lower limit value The peak value stored in the peak value storage means and the peak value of the volume level measured by the measurement means when there is a difference of a predetermined magnitude or more, the peak value storage means a second judging means judges that updating of said stored peak value, updating said peak value by the first determination unit and the second judging means An update means for updating the peak value stored in the peak value storage means with the peak value measured by the measurement means, and the peak value storage means updated by the update means, As a signal processing determined according to the peak value for the audio signal transmitted to the communication destination device based on the peak value of the volume level stored in the volume level, the volume level is determined according to the peak value. Signal processing means for performing at least one of a process for changing to a level determined and a process for changing the level of echo cancellation to a level determined in accordance with the peak value is provided.

  In this audio processing device, the signal processing means performs peak processing based on the peak value of the volume level measured by the measurement means when performing signal processing on the audio signal transmitted to the communication destination device. The signal processing determined according to is performed. Such a peak value of the sound volume level can be measured almost instantaneously, unlike an integrated value over a predetermined period.

Therefore, with this audio processing device, for example, even when the volume changes frequently, such as inflected audio, signal processing can be performed without delaying the actual change in volume, making the volume easier to hear. And sound quality can be corrected.
Further, the first determination unit determines to update the peak value stored in the peak value storage unit when the peak value of the volume level measured by the measurement unit is larger than a predetermined lower limit value. To do.
Therefore, according to the sound processing device configured as described above, even if there is a change in the peak value of the volume level measured by the measuring unit, the first determination unit stores the peak value stored in the peak value storage unit. Only when it is determined to update, the updating means does not update the peak value stored in the peak value storing means, and in particular, the peak value of the volume level measured by the measuring means is below a predetermined lower limit value. If there is, the peak value stored in the peak value storage means is not updated.
Therefore, when the peak value becomes extremely small, the content of the signal processing by the signal processing means does not change, so that the volume is not increased in accordance with the instantaneous silent state, and the user can not hear it. You can prevent the feeling of unnaturalness.

In the audio processing device according to the present invention, the signal processing means may change a volume level to a level determined according to the peak value, and an echo cancellation level as signal processing determined according to the peak value. of the process of changing a level determined in accordance with the peak value, that is configured to at least one treatment, applied to the audio signals transmitted with the communication destination device.

  If the process of changing the volume level to a level determined according to the peak value is performed, for example, the volume is reduced as the peak value is larger, or the volume is increased as the peak value is smaller. It can make it easy to hear.

  As for the echo cancellation processing, the echo (the bounce sound of your voice) tends to increase as the peak value increases, so if the peak value is large, the echo cancellation works more strongly, while the peak value is By changing the echo cancellation level (setting) so that the echo cancellation works more weakly when it is small, it is possible to make it easier to hear the voice. By the way, if the echo is excessively large, the voice of the other party tends to be difficult to hear, but even if the echo is completely lost, there is a sense of incongruity, and there is a tendency that it is not always easy to hear and speak. That is, it is a problem if the echo cancellation is too weak, but it is not ideal that the echo cancellation is too strong. In this regard, if the echo cancellation level is changed according to the peak value as described above, the echo will not be excessively increased when the volume is high, and the echo will not be excessively decreased when the volume is low. .

These treatments can achieve a reasonable effect by adopting only one, but are more effective if both are adopted .

In the audio processing device of the present invention , the second determination means has a predetermined magnitude between the peak value stored in the peak value storage means and the peak value of the volume level measured by the measurement means. If there is a difference greater than or equal to this, it is determined to update the peak value stored in the peak value storage means, and the update means is determined to update the peak value by the second determination means. when the, the peak value stored in the peak value storage unit, to update with the peak value measured by said measuring means.

Therefore , since the content of the signal processing by the signal processing means does not change when the peak value is slightly changed, it is possible to prevent the user from feeling unnatural on hearing.

The above-described voice processing apparatus of the present invention is preferably applied to a call device such as an IP phone device configured using a telephone, a personal computer, or the like.
Further, if there is a program for causing a computer or personal computer incorporated in the telephone to function as each means included in the above-described voice processing apparatus, the voice processing apparatus of the present invention can be configured using those computers. it can.

Next, an embodiment of the present invention will be described with an example.
In the embodiments described below, the voice processing apparatus according to the present invention is called a multi-function product (generally MFP (Multi-Function Product)) having a facsimile function, a telephone function, a printer function, a scanner function, and a copy function. Device). The voice processing apparatus according to the present invention performs signal processing, which will be described later, on a voice signal transmitted to a communication destination device when the telephone function of the multifunction machine is used.

[Configuration of MFP]
First, the configuration of the multifunction machine 1 will be described.
As shown in FIG. 1, the multifunction device 1 includes a CPU 11, a ROM 12, an EEPROM 13, a RAM 14, an image memory 15, a line I / F unit 19, a modem 20, a buffer 21, a scanner 22, an encoding unit 23, and a decoding unit 24. , A printer 25, an operation panel 4, an LCD (liquid crystal display panel) 5, an amplifier 27, a telephone handset 47, and the like, which are connected to each other via a bus line 30.

The CPU 11 is a device that controls each unit connected by the bus line 30 and executes, for example, processing for transmitting and receiving voice and image data via the line I / F unit 19.
The ROM 12 is a non-rewritable storage device that stores a control program executed by the multifunction device 1. A program for causing the CPU 11 to execute processing to be described later is stored in the ROM 12. The ROM 12 also stores a handset acoustic parameter table 12a (details will be described later).

The EEPROM 13 is a rewritable nonvolatile storage device that can retain written data even after the power of the multifunction device 1 is shut off.
The RAM 14 is a storage device that can be read and rewritten to store various data, incoming call history, and the like when each operation of the multifunction device 1 is executed, and is a handset acoustic parameter that is written and updated during processing to be described later. 14a, the peak audio level 14b, and the like are stored in the RAM 14.

  The image memory 15 is a storage device for storing image data and a bit image for printing, and includes a dynamic RAM (DRAM) that is an inexpensive large-capacity memory. The received image data is temporarily stored in the image memory 15, printed on recording paper by the printer 25, and then erased from the image memory 15. The image data read by the scanner 22 is also stored in the image memory 15.

  The line I / F unit 19 is used for line control, such as a call signal (ring signal) sent from a telephone line side (for example, an exchange or an IP telephone adapter), a telephone number of a counterpart device, etc. Various signals such as a signal indicating the sender identification information (hereinafter referred to as “CallerID”) are received and a dial signal at the time of outgoing call according to the operation of the key on the operation panel 4 is transmitted to the telephone line side. To do.

The modem 20 is used to modulate and demodulate and transmit image information and communication data, and to transmit and receive various procedure signals for transmission control.
The buffer 21 is for temporarily storing data including encoded image information transmitted / received to / from the counterpart device.

The scanner 22 is for reading a document inserted into the document insertion slot as image data, and includes a document transport motor.
The encoding unit 23 encodes image data read by the scanner 22.

  The decoding unit 24 reads image data stored in the buffer 21 or the image memory 15 and decodes it. The decoded data is printed on a recording sheet by the printer 25.

  The printer 25 is a well-known inkjet printer including a recording paper transport motor that transports recording paper, a carriage motor that moves a carriage on which the print head is mounted, and a print head that ejects ink onto the recording paper. Has been.

  The LCD 5 emits various information to the user through characters or images based on a command signal from the CPU 11. The information includes, for example, the fact that FAX data is being transmitted / received or the telephone number of the caller Or the information of the sender such as a facsimile number and the name or name of the sender can be used.

The amplifier 27 is used to ring a speaker 28 connected to the amplifier 27 and output a ringing tone or voice.
The handset 47 is a handset for a call in which a speaker that reproduces a sound signal transmitted from a communication destination device and a microphone that inputs sound emitted from a caller are integrated.

  The hook switch 48 is turned on when the user performs an operation (hook up) for lifting the handset 47, and is turned off when an operation for returning the handset 47 to the original position (hook down) is performed. . On / off of the hook switch 48 is monitored by the CPU 11, and when the hook switch 48 is turned on, the CPU 11 closes the line, and when the hook switch 48 is turned off, the CPU 11 opens the line. Further, the hook switch 48 can be switched on / off by an operation on the operation panel 4. If the operation panel 4 is used, an operation (hook up) for lifting the handset 47 is not actually performed. In addition, the state of the multifunction device 1 can be switched to a state equivalent to the case where hookup / hookdown is performed.

[Handset acoustic parameter table]
Next, the handset acoustic parameter table 12a stored in the ROM 12 will be described.

  In the present embodiment, the handset acoustic parameter table 12a has a data structure as shown in FIG. In other words, in the present embodiment, three sets of acoustic parameter groups are stored in the handset acoustic parameter table 12a, and these three sets of acoustic parameter groups have three ranges (−20 dBm to −30 dBm, −10 dBm to Corresponding in a one-to-one relationship with the signal level (= volume level) of the audio signal divided into −20 dBm and −10 dBm or more).

  Each set of acoustic parameter groups includes a transmission level, a reception level, a filter characteristic value, an LEC (line echo canceller) setting value, and the above four types of acoustic parameters. Among these four types of acoustic parameters, for two types of transmission level and reception level, there are three pairs of acoustic parameters corresponding to each of the three volume settings (large, medium, and small) that can be set on the operation panel 4. Stored.

  Among the three acoustic parameter groups, the acoustic parameter group corresponding to −20 dBm to −30 dBm is a parameter for performing signal processing on a relatively small sound, and the signal processing is performed based on the acoustic parameter group. As a result, volume is increased, noise components are cut, echo cancellation, and the like are performed, and a relatively small sound is converted into a volume and sound quality that are easier to hear.

  Also, the acoustic parameter group corresponding to −10 dBm to −20 dBm is a parameter for performing signal processing on a sound with a standard volume, and by performing signal processing based on this acoustic parameter group, No change is made, noise components are cut, echo cancellation, etc. are performed, and the sound of standard volume is converted to a sound quality that is easier to hear.

  Furthermore, the acoustic parameter group corresponding to -10 dBm or more is a parameter for performing signal processing on a relatively loud sound. By performing signal processing based on this acoustic parameter group, sound volume reduction, noise The components are cut, echo cancelled, etc., and a relatively loud sound is converted into a volume and sound quality that are easy to hear.

  Of these three acoustic parameter groups, any one of the groups will be used alternatively in the processing to be described later, and signal processing is performed on the audio signal transmitted to the communication destination device. Become. At this time, for the two types of transmission level and reception level, the volume setting (high volume) set on the operation panel 4 is selected from the acoustic parameters for which three pairs corresponding to the volume setting (high, medium, low) are prepared. , Medium, and small), any one pair of acoustic parameters is used.

  In the processing to be described later, when the CPU 11 measures the peak value of the signal level of the audio signal transmitted from the communication destination device and satisfies a predetermined update condition (described later in detail), the measured peak value is Stored in the peak audio level 14b. When the peak value stored in the peak audio level 14b is updated, the volume setting (large, medium, small) on the operation panel 4 is selected from the set of acoustic parameters corresponding to the peak value. In consideration, the four types of acoustic parameters are read out, and the read out four types of acoustic parameters are stored in the handset acoustic parameter 14a.

  The CPU 11 is configured to always perform signal processing with reference to the handset acoustic parameter 14a for the audio signal transmitted to the communication destination device, and the peak value stored in the peak audio level 14b is updated. When this is done, the handset acoustic parameter 14a is updated along with the update. As a result, the CPU 11 responds to the updated peak value with respect to the audio signal transmitted to the communication destination device. The determined signal processing is performed.

[Signal processing for audio signals]
Next, signal processing for an audio signal will be described based on the flowchart shown in FIG. The flowchart shown in FIG. 3 shows only the processing steps related to the main part of the present invention extracted from the processes that are always executed by the CPU 11 in the multifunction machine 1. The CPU 11 actually executes processing that does not appear in FIG. 3, but illustration of processing that is not related to the main part of the present invention is omitted.

  When this process is executed by the CPU 11, first, the hook switch 48 is monitored (S105), and it is determined whether or not the user has hooked up the handset (S110). Here, when the user has not hooked up the handset (S110: No), the process returns to S105, the processes of S105 to S110 are repeated, and the monitoring of the hook switch 48 is continued. In the process of S110, not only when the user actually hooks up the handset 47, but also when the state of the multifunction device 1 is switched to the same state as when the state of the multifunction device 1 is hooked up by the operation on the operation panel 4. It is determined that the handset is hooked up.

  Then, if the user hooks up the handset while the processes of S105 to S110 are repeated (S110: Yes), the line is closed (S115), and the call start process by the handset 47 is executed (S120). The person can enter a call with the other party using the communication destination device.

  Subsequently, the peak value (Lp) is initialized (S125). This peak value (Lp) is a value stored in the peak audio level 14b of the RAM 14, and the signal level (volume level) of the audio signal transmitted to the communication destination device is measured in the process described later. After that, the peak value of the measured signal level is stored, but since the signal level of the audio signal transmitted to the communication destination device has not been actually measured at the stage of processing of S125. In the process of S125, 0 dBm is stored in the peak audio level 14b of the RAM 14 as an initial value.

  Subsequently, initialization of handset acoustic parameters is performed (S130). In the process of S130, the acoustic parameter group corresponding to the initial value (0 dBm) of the peak value (Lp) is read from the handset acoustic parameter table 12a of the ROM 12. At this time, the volume setting (high, medium, small) set on the operation panel 4 is also referred to, and two types of acoustic parameters (sending level and receiving level) according to the volume setting and the other two types of sound are set. Parameters (filter characteristic value and LEC set value) are read. Then, the read four kinds of acoustic parameters are stored in the handset acoustic parameter 14 a of the RAM 14. As already described, since the CPU 11 always performs signal processing on the audio signal transmitted to the communication destination device with reference to the handset acoustic parameter 14a, the communication is performed when the processing of S130 is performed. Signal processing is performed on the assumption that the peak value of the signal level of the audio signal transmitted to the destination device is 0 dBm.

  As described above, the reason for performing signal processing on the assumption that the peak value of the signal level of the audio signal is 0 dBm is that the signal level of the audio signal actually transmitted to the communication destination device is relatively excessive. If it is, the volume can be reduced appropriately, and even if the signal level is relatively low, the volume is only further reduced. This is because it is possible to prevent the voice from being emitted. If signal processing is performed on the assumption that the peak value of the signal level of the audio signal transmitted to the communication destination device is −30 dBm, the audio signal actually transmitted to the communication destination device If the signal level is relatively low, the volume may be increased appropriately, but if the signal level is relatively high, the volume will increase further. There is a risk of unexpected loud sound from the handset 47.

  When the above processing is completed, the hook switch 48 is monitored (S135), and it is determined whether or not the handset is hooked down (S140). If it is determined that the handset is not hooked down (S140: No), the signal level (Lc) of the audio signal transmitted to the communication destination device is measured (S145).

  The absolute value (| Lp−Lc |) of the difference between the peak value (Lp) stored in the peak audio level 14b of the RAM 14 and the signal level (Lc) measured in the process of S145 is a predetermined threshold value. It is determined whether or not it is greater than Lh (5 dBm in this embodiment) (S150). If | Lp-Lc |> Lh (S150: Yes), the signal level (Lc) measured in the process of S145 is −. It is determined whether it is larger than 30 dBm (S155).

  Here, if Lc> −30 dBm (S155: Yes), the peak audio level is updated (S160). Specifically, the peak value (Lp) is updated by storing the signal level (Lc) measured in the process of S145 in the peak audio level 14b of the RAM 14. And the handset acoustic parameter according to this new peak value (Lp) is set (S165). In the process of S165, four types of acoustic parameters (transmission level, reception level, filter characteristic value, and LEC setting value) corresponding to the new peak value (Lp) are read from the handset acoustic parameter table 12a of the ROM 12. Then, the read acoustic parameter group is stored in the handset acoustic parameter 14 a of the RAM 14. Thereafter, since the CPU 11 performs signal processing on the audio signal transmitted to the communication destination device with reference to the handset acoustic parameter 14a, signal processing based on a new acoustic parameter group is performed. When the process of S165 is completed, the process returns to S135.

  On the other hand, if | Lp−Lc | ≦ Lh in the process of S150 (S150: No), the process returns to S135 without executing the processes of S160 to S165. By performing such a process, the signal level (Lc) measured in the process of S145 is the boundary value of the range of the three signal levels associated with the three acoustic parameter groups (in the case of the present embodiment). When the frequency fluctuates in the vicinity of −10 dBm, −20 dBm), it is possible to prevent the acoustic parameter group from being updated excessively frequently. To give a specific example, for example, when the signal level (Lc) measured in the process of S145 varies between −9 dBm to −11 dBm across the boundary value of −10 dBm, the process of S150 is performed. Otherwise, the update of the acoustic parameter group is performed excessively frequently, and as a result, there is a possibility that the volume reduction / reduction stop is frequently repeated. However, if the process of S150 is performed, the acoustic parameter group is updated. Since the update is not performed, the volume reduction / reduction stop is not frequently repeated.

  Also, in the process of S155, if Lc ≦ −30 dBm (S155: No), the process returns to S135 without executing the processes of S160 to S165. By performing such processing, the acoustic parameter group is updated only when the signal level (Lc) measured in the processing of S145 instantaneously becomes silent or nearly silent, and an unnecessary increase in volume is achieved. It can be prevented from being planned.

  As described above, when a negative determination is made in the process of S150, a negative determination is made in the process of S155, or when the process of S165 is completed, the process returns to the process of S135 in any case. As long as it is determined that the handset is not hooked down in the process (S140: No), the processes in S135 to S165 are repeated. That is, during the call, the processes of S135 to S165 are repeated. Then, in the repetition processing of S135 to S165, the absolute value (| Lp) of the difference between the peak value (Lp) stored in the peak audio level 14b of the RAM 14 and the signal level (Lc) measured in the processing of S145. -Lc |) is larger than a predetermined threshold Lh (S150: Yes) and the signal level (Lc) measured in the process of S145 is larger than -30 dBm (S155: Yes), the acoustic parameter The group is updated, and as a result, the audio signal transmitted to and from the communication destination device is easy to hear because the volume is optimized, the noise component is cut according to the volume, the echo is canceled according to the volume, etc. It is corrected to.

  If it is determined that the handset is hooked down in the process of S140 during the repeat process of S135 to S165 as described above (S140: Yes), the line is released (S170), and the process returns to S105. . Thereafter, the processing of S105 to S110 is repeated until it is determined that the handset is hooked up again.

In the embodiment of the present invention described above, the CPU 11 that executes the process of S145 corresponds to the measurement means referred to in the present invention, and refers to the acoustic parameter group stored in the handset acoustic parameter 14a of the RAM 14 by the process of S165. And CPU11 which performs signal processing with respect to the audio | voice signal transmitted between communication destination apparatuses is equivalent to the signal processing means said by this invention. The RAM 14 (peak audio level 14b) in which the signal level is stored by the process of S160 corresponds to the peak value storage means in the present invention, and the CPU 11 that executes the process of S155 is referred to in the present invention. The CPU 11 that corresponds to the first determination means and executes the process of S150 corresponds to the second determination means in the present invention, and the CPU 11 that executes the process of S160 corresponds to the update means in the present invention.

As mentioned above, although embodiment of this invention was described, this invention is not limited to said specific one Embodiment, In addition, it can implement with a various form.
For example, in the above-described embodiment, an example in which the voice processing device of the present invention is applied to the multi-function device 1 has been described. Specifically, the configuration of the present invention may be adopted in a single function telephone having no facsimile function or the like. In addition to general telephones connected to the PSTN network, telephones include telephones connected to the IP network (so-called IP telephones) and mobile phones. The configuration of the present invention is applicable to any of these telephones. Can be adopted. Furthermore, if a microphone and headphones are connected to a personal computer and IP telephone software is installed in the personal computer, an IP call using the personal computer can be performed. The personal computer is configured using such a personal computer. The configuration of the present invention can also be adopted in an IP telephone device.

  In the above embodiment, the signal level of the audio signal is divided into three ranges, and three sets of acoustic parameter groups corresponding to each range are stored in the handset acoustic parameter table 12a. May not necessarily be divided into three ranges, for example, it may be divided into two ranges or four or more ranges. Further, the boundary value of each range is not limited to −10 dBm and −20 dBm, and can be arbitrarily changed. Furthermore, although one set of acoustic parameter groups is composed of four types of acoustic parameters, the number of acoustic parameters is also arbitrary.

  In the above embodiment, the acoustic parameter group is read from the handset acoustic parameter table 12a prepared in advance. However, the method for obtaining the acoustic parameter group is not limited to the table reference method. Instead, for example, a predetermined acoustic parameter y = f (x) may be calculated using a predetermined mathematical formula f (x) having the signal level x of the audio signal as a variable.

  In the above embodiment, the threshold value Lh is set to 5 dBm in the process of S150. However, a value other than 5 dBm may be adopted for the threshold value Lh. In the process of S155, the signal level ( Although it has been determined whether or not Lc) is greater than −30 dBm, a value other than −30 dBm may be employed for this value.

  Furthermore, there are two audio signals transmitted to and from the communication destination device: an audio signal transmitted to the communication destination device and an audio signal transmitted from the communication destination device, both of these two audio signals. Is the most effective processing target by the speech processing apparatus of the present invention, but the corresponding effect can be obtained even if only one of the speech signals is processed by the speech processing apparatus of the present invention. That is, the configuration of the present invention may be applied to both the transmission side and the reception side, may be applied only to the transmission side, or may be applied only to the reception side. When applied to the transmitting side, even if the voice emitted by the user of this communication device is too low or too high, it can be transmitted to the destination device after being corrected to an appropriate volume and sound quality. When applied to the receiving side Can correct the sound and the sound quality to be appropriate even if the sound transmitted from the communication destination device is too small or too large, and then can hear the sound after the correction to the user of the call device.

The block diagram which shows schematic structure of the telephone apparatus demonstrated as embodiment of this invention. Explanatory drawing which shows an example of a handset acoustic parameter table. The flowchart which shows the signal processing with respect to an audio | voice signal.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... MFP, 4 ... Operation panel, 5 ... LCD, 11 ... CPU, 12 ... ROM, 12a ... Handset acoustic parameter table, 13 ... EEPROM, 14 ... RAM, 14a: handset acoustic parameters, 14b: peak sound level, 15: image memory, 19: line I / F unit, 20: modem, 21: buffer, 22 ..Scanner, 23 ... encoding unit, 24 ... decoding unit, 25 ... printer, 27 ... amplifier, 28 ... speaker, 30 ... bus line, 47 ... handset 48 ... Hook switch.

Claims (3)

Measuring means for measuring the volume level of the audio signal transmitted from the communication destination device;
Peak value storage means capable of storing a peak value of the volume level measured by the measurement means;
A first determination unit that determines to update the peak value stored in the peak value storage unit when a peak value of the volume level measured by the measurement unit is greater than a predetermined lower limit value; ,
When the peak value stored in the peak value storage means and the peak value of the volume level measured by the measurement means have a difference greater than a predetermined magnitude, the peak value storage means stores the difference. Second determination means for determining to update the peak value,
When the peak value stored in the peak value storage unit is determined to be updated by the first determination unit and the second determination unit , the peak value measured by the measurement unit is used. Update means to update with,
Based on the peak value of the volume level stored in the peak value storage unit updated by the updating unit, a signal determined according to the peak value with respect to an audio signal transmitted to the communication destination device Signal processing means for performing at least one of processing for changing the volume level to a level determined according to the peak value and processing for changing the level of echo cancellation to a level determined according to the peak value as processing A voice processing apparatus comprising: A communication device comprising the voice processing device according to claim 1 . The program for functioning a computer as each means with which the audio | voice processing apparatus of Claim 1 is provided.

Images