JP2003044075A - Electronic equipment with speech recognizing function - Google Patents

Abstract

PROBLEM TO BE SOLVED: To solve such a problem that, when electronic equipment outputting voice is operated by voice, conventional equipment has lacked swiftness because it has performed time-consuming recognition processing even when a voice signal inputted is obviously unsuitable for recognition processing before performing it. SOLUTION: The electronic equipment is provided with a voice level detection part 104, a voice section detection part 106, and a loudness control part 111, and when the voice level detection part 104 judges that an input level is too high, and the voice range detection part 106 judges that a voice section is too long, the loudness control part 111 is informed to reduce loudness of the voice signal outputted from the electronic equipment before a word recognition part 107 performs recognition processing.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electronic device having a voice recognition function, and more particularly to an electronic device which outputs a voice from an electronic device main body.

[0002]

2. Description of the Related Art In recent years, a voice interface for operating electronic equipment by voice has been realized.

This is because the functions of electronic devices have become complicated and multifunctional, and the switches and buttons for operating have become diversified, so that the operation procedure of the devices becomes complicated and the intended operation is completed. Because it was necessary to perform button operations more than once, it was possible to improve the user interface by using voice recognition technology, such as enabling these operations with a single voice command.

However, when an electronic device that realizes a voice interface outputs a voice signal from its main body such as a television or an audio device, this output voice is input together with a voice command uttered by a speaker, There is a problem that it becomes noise when performing the recognition process, recognition failure occurs, and the recognition performance deteriorates.

In such a case, that is, in the case where the recognition performance is deteriorated by the influence of a voice signal other than the utterance of the speaker (referred to as the surrounding environment of the utterance), an apparatus for performing an improvement process for improving the surrounding environment is provided. It is disclosed.

[0006] For example, Japanese Patent Laid-Open No. 11-126092 discloses a voice recognition device for improving the surrounding environment when a speaker's voice recognition fails due to the surrounding environment.

FIG. 3 shows a block diagram of a conventional voice recognition apparatus.

In FIG. 3, reference numeral 201 denotes a voice input unit, which inputs a voice uttered by a speaker. A voice recognition unit 202 recognizes voice data input from the voice input unit 201 and outputs a recognition result. A device control unit 203 controls each device of the audio device 204 and the air conditioner 205 based on the recognition result obtained by the voice recognition unit 202.

In the conventional speech recognition apparatus configured as described above, when the speech recognition apparatus fails to recognize the speech of the speaker due to the surrounding environment, the operation when the speech recognition apparatus improves the surrounding environment will be described below. explain.

When the voice from the speaker is input, the voice input unit 201 converts the input voice data into a digital signal and outputs the digital signal to the voice recognition unit 202. Voice recognition unit 202
Performs a recognition process of comparing the voice data input from the voice input unit 201 with the standard voice data. Speech recognition unit 2
When the recognition process fails, 02 sends a request for improving the surrounding environment of the voice input unit 201 to the device control unit 203. When a request for improving the surrounding environment of the voice input unit 201 is sent from the voice recognition unit 202, the device control unit 203 outputs a control signal to the control target device, that is, the audio device 204 and the air conditioner 205, and controls the device. To operate. For example, when the causal device is the audio device 204, the volume of the audio device 204 is lowered. As a result, the environment around the voice input unit 201 is improved, and the recognition performance for the next utterance of the speaker can be improved.

[0011]

However, in the above configuration, when the voice level or voice section of the voice data uttered by the speaker is clearly inappropriate for the recognition process, that is,
Even if there is a high possibility that the voice recognition unit will not perform the recognition process correctly, input the voice data to the voice recognition unit to perform the time-consuming recognition process, and improve the surrounding environment of the utterance after the recognition fails. Therefore, there is a problem that the operability of the device is deteriorated such that the speaker cannot immediately re-voice.

Therefore, when the input voice data is obviously inappropriate for the recognition process, it is necessary to quickly control the voice output from the electronic equipment body without performing the time-consuming recognition process. is there.

The present invention has been made in order to solve the above problems, and an object of the present invention is to provide an electronic device which outputs voice, and when the electronic device is operated by voice, the input voice data is obviously unclear to the recognition process. When appropriate, control the volume of the audio signal output from the electronic device itself, and quickly improve the surrounding environment for the speaker's utterance without time-consuming recognition processing to operate the electronic device. Is to improve the sex.

[0014]

[Means for Solving the Problems] The first invention (Claim 1)
Corresponds to), in an electronic device equipped with a voice recognition means, a voice output means for outputting a voice, and a volume control means for controlling the volume of the voice output means, to detect the start of a voice command of a speaker. A voice detection unit, a voice collection unit that collects a voice, a voice input level detection unit that detects a level of the collected voice and controls the volume control unit based on the level, and a voice input level detection unit that obtains from the voice recognition unit. The electronic device is provided with at least a recognition result determination means for outputting a voice output control signal for executing the voice command of the speaker based on the recognized recognition result.

In a second aspect of the present invention (corresponding to claim 2), the voice input level detecting means holds a predetermined level parameter so as to correspond to the level of a voice command of an input voice, If the audio level is higher than the level parameter, the electronic device according to the first aspect of the present invention outputs a volume control signal for reducing the volume of audio to the volume control means.

According to a third aspect of the present invention (corresponding to claim 3), an electronic device comprising voice recognition means, voice output means for outputting voice, and volume control means for controlling the volume of the voice output means. In, a voice detecting means for detecting the start of a voice command of a speaker, a voice collecting means for collecting voice, a voice section of the collected voice is detected, and the volume control is performed based on the length of the voice section. And a recognition result determining means for outputting at least a voice output control signal for executing the voice command of the speaker based on the recognition result obtained from the voice recognition means. It is an electronic device.

In a fourth aspect of the present invention (corresponding to claim 4), the voice section detecting means holds a predetermined section detection parameter so as to correspond to a voice section of a voice command of an input voice. If the voice section is longer than the section detection, the electronic device according to the third aspect of the present invention outputs a volume control signal for reducing the volume of voice to the volume control unit.

According to a fifth aspect of the present invention (corresponding to claim 5), the voice detecting means is a sensor for detecting that a person has entered a predetermined area with the voice collecting means as a reference, or a speaker himself / herself. The electronic device according to the first or third aspect of the present invention, which comprises a switch or a sensor capable of notifying that a voice command is about to start by a method other than voice.

A sixth aspect of the present invention (corresponding to claim 6) is based on volume setting value detecting means for detecting the setting value of the volume output by the device, and the setting value detected by the volume setting value detecting means. The electronic device according to the first or third aspect of the present invention is provided with a parameter changing unit that indicates changing the level parameter or the section detection parameter.

The seventh invention (corresponding to claim 7)
Is a voice detecting means for detecting the start of a voice command of a speaker of the electronic device of the first aspect of the present invention, detects the level of collected voice, and controls the volume control means based on the level. As a whole or a part of the voice input level detecting means and at least the recognition result determining means for outputting a voice output control signal for executing the voice command of the speaker based on the recognition result obtained from the voice recognizing means. A program for operating a computer.

The eighth invention (corresponding to claim 8)
Is a voice detecting means of the electronic device of the third aspect of the invention for detecting the start of a voice command of a speaker; and a voice section of the collected voice, and based on the length of the voice section, the sound volume. Based on the recognition result obtained from the voice section detection means for controlling the control means and the voice recognition means, at least,
It is a program for causing a computer to function as a whole or a part of a recognition result determining unit that outputs a voice output control signal for executing a voice command of the speaker.

The ninth invention (corresponding to claim 9)
Is a voice detecting means for detecting the start of a voice command of a speaker of the electronic device of the first aspect of the present invention, detects the level of collected voice, and controls the volume control means based on the level. As a whole or a part of the voice input level detecting means and at least the recognition result determining means for outputting a voice output control signal for executing the voice command of the speaker based on the recognition result obtained from the voice recognizing means. A medium carrying a program for causing a computer to function, which is characterized by being processable by the computer.

The tenth aspect of the present invention (corresponding to claim 10) is a voice detecting means for detecting the start of a voice command of a speaker of the electronic device according to the third aspect of the invention, and a voice of collected voice. At least a voice command of the speaker is detected based on a voice period detection unit that detects a period and controls the volume control unit based on the length of the voice period, and a recognition result obtained from the voice recognition unit. A medium carrying a program for causing a computer to function as all or a part of a recognition result determination means for outputting a voice output control signal for executing the above, and characterized by being processable by the computer.

[0024]

BEST MODE FOR CARRYING OUT THE INVENTION Regarding one embodiment of the present invention,
This will be described with reference to the drawings.

Regarding the configuration of the electronic equipment of the present embodiment,
This will be described with reference to FIG.

As shown in FIG. 1, the present electronic device has a voice input unit 101, a voice recognition unit 105, and a recognition result determination unit 1
08, a voice output control unit 109, and a voice output unit 113
And a video output control unit 114 and a video output unit 115.

The voice input unit 101 includes a voice detection unit 10
2, a microphone 103, and a voice input level detection unit 104.

The voice detection unit 102 is means for detecting the start of utterance of a voice command uttered by a speaker to operate an electronic device by a predetermined method.

The voice detecting section 102 is, for example, a human sensor, and when a person enters the detection area of the sensor, the sensor operates by sensing the heat (infrared ray) generated by the person. As a result, it is detected that the person approaches the voice input unit 101. A touch sensor or the like, which is a part of the voice input unit or the voice input unit 101 before the speaker starts speaking.
The sensor is activated by touching a part of the electronic equipment other than, and detects the speaker. The voice detection unit 102 may be a talk switch, and detects the start of the speaker's utterance by pressing the switch before the speaker speaks. The talk switch includes a press talk switch that presses the switch continuously to make a voice and a push talk switch that presses the switch before starting to make a voice. Either of them may be used.

The microphone 103 is a voice collecting device,
The voice uttered by the speaker is input.

The voice input level detecting section 104 includes a microphone 1
The voice level of the voice input to 03 is detected. If the voice level of the input voice is higher than a predetermined level parameter (α, α: real number), a volume control signal is output to the volume control unit 111. If the voice level of the input voice is lower than α, the input voice is output to the voice section detection unit 104 as voice data. In other words, when the volume is large, a voice of a person or the like is output from the electronic device, it is determined that the voices are duplicated, and a volume control signal for reducing the voice output of the electronic device is output.

When the parameter change notification signal is input from the parameter change notification unit 112, the voice input level detection unit 104 changes the value of the level parameter α based on the content of the signal.

The voice recognition unit 105 includes a voice section detection unit 10
6 and a word recognition unit 107.

The voice section detector 106 detects the voice section using the voice level of the voice data output from the voice input level detector 104 and the result of frequency analysis of the voice data. If the voice section of the voice data is longer than a predetermined section detection parameter (t, t: real number),
A volume control signal is output to the volume control unit 111. If the voice section of the voice data is shorter than the section detection parameter t, the voice data is output to the word recognition unit 107. That is, there is an empirical rule that the length of a command command by voice is less than or equal to a certain value, and the fact that the voice section is longer than the certain value means that the voice input means is separated from the sound generated by the speaker in the electronic device. Since it can be estimated that the voices output from the are input at the same time, the voice output of the electronic device is reduced.

When the parameter change notification signal is input from the parameter change notification unit 112, the voice section detection unit 106 changes the value of the section detection parameter t based on the content of the signal.

The word recognition unit 107 holds a recognition dictionary (not shown) in which usable voice commands are stored as standard voice data, and compares the voice data with the standard voice data by using the recognition dictionary. Perform recognition processing. Regarding the recognition processing, for example, as shown in "Voice Communication with Computer-For Future Computing-" (Science) by C. Schmann, the word recognition unit 107 is input from the voice section detection unit 106. A frame having an LPC coefficient every 20 milliseconds is extracted from the audio data by digital signal processing. Next, the frame of the obtained LPC coefficient is compared with the LPC coefficient data of each voice command held in the recognition dictionary as standard voice data by using a matching algorithm. The matching algorithm calculates, for example, for each dimensional vector of the LPC coefficient, the distance from the vector of each dimension of the standard voice data, calculates the sum of the distances, and sets this as the distance between the input voice data and the standard voice data. Next, when the distance between the input voice data and the standard voice data is the smallest, it is normalized so that it becomes the highest point (for example, 100), and this is taken as the reliability of the recognition word, and the reliability of the recognition word is the highest. Standard speech data is obtained, and the words and the reliability of the words are set as one set, and a predetermined number of sets are output as recognition results.

The recognition result determination unit 108 is the word recognition unit 10.
The recognition success / failure is determined based on the recognition word output from 7 and the reliability of the word. Rejection parameter with predetermined word reliability (φ, φ: real number)
If it is larger, it is determined that the recognition is successful, the content of the recognized word is analyzed, and a voice output control signal is output to the voice output control unit 109 or a video output control signal is output to the video output control unit 114. If the reliability of the recognized word is smaller than the reject parameter φ, it is determined that the recognition has failed and the recognized word is rejected. Then, the recognition failure outputs a sound volume control signal to the sound volume control unit 111 so as to estimate that sound from the electronic device is simultaneously output and reduce the sound.

The audio output control unit 109 includes a channel switching unit 110, a volume control unit 111, and a parameter change notifying unit 1.
It is composed of 12 and.

For example, in the case of a television, the channel switching section 110 switches the channel of the television sound. When the speech output control signal is input as a result of the successful recognition, the speech output section 113 is based on the content thereof. Switch the audio source channel to be output to.

The volume control unit 111 includes the units 104 and 10
When the volume control signals are input from the voice control unit 6, 108, etc., the audio output unit 1 is in accordance with the contents of the volume control signals.
The volume of the audio signal output from 13 is controlled.

When the set value of the volume of the audio signal output from the audio output unit 113 is changed, the parameter change notification unit 112 sends the value of the level parameter α to the audio input level detection unit 104 based on the set value of the volume. And outputs a parameter change notification signal indicating that the parameter is changed. Alternatively, a parameter change notification signal indicating that the value of the section detection parameter t is changed based on the volume setting value is output to the voice section detection unit 106. The parameter change notification unit 112 outputs the volume control signal before performing the volume control, for example, when the power of the electronic device is turned on or when the channel switching unit 110 changes the channel of the audio source output to the audio output unit 113. The sound volume control unit 111 outputs the sound output unit 11
Before controlling the volume of the sound output from the audio output unit 3, both or one of the level parameter α and the section detection parameter t is changed based on the volume setting value set by the volume control unit 111. If the level parameter α and the section detection parameter t have already been set to appropriate values, they need not be changed.

That is, the parameters are not changed during the recognition, and when the power of the electronic device is turned on, the recognition is successful and the electronic device is operated. As a result, the volume output from the electronic device is significantly changed. In such a case, if the same parameters as before are used, it becomes impossible to make an accurate judgment, and therefore the parameters are changed.

The voice output unit 113 is a voice output device such as a speaker, and outputs the voice signal obtained from the voice output control unit 109.

The video output control unit 114, for example, in the case of a television, switches the channels of television video.
When the video output control signal is input, the channel of the video source to be output to the video output unit 115 is switched based on its content.

The video output unit 115 is a video display device such as a display, and outputs the video signal obtained from the video output control unit 114.

The operation of the electronic device configured as described above will be described below with reference to the flowchart of FIG.

In the voice input waiting state S300, the electronic device is in a state of waiting for voice input from the speaker. At this time, the level parameter α and the section detection parameter t
Is set to an appropriate value based on the parameter change notification signal output by the parameter change notification unit 112 based on the set value of the volume of the audio signal output from the audio output unit 113.

In step S301, the voice detection unit 102
Detects the start of the speaker's utterance and determines that the speaker has started uttering a voice command for operating the electronic device, the speaker's voice is input to the microphone 103, and step S
Transition to 302.

The method for detecting the start of the speaker's utterance by the voice detecting unit 102 is, for example, using a push-talk switch. When the speaker presses the push-talk switch, the talker's utterance start is detected and The voice command is input to the microphone. When it is not possible to detect the start of speaking by the speaker, that is, when the push-talk switch is used and the switch is not pressed, a voice input waiting state S30
0 continues.

At step S302, the microphone 103
The input voice is output as voice data to the voice input level detection unit 104, and the voice input level detection unit 104 outputs the voice level of the voice data output from the microphone 103 and α
Compare with. If the voice level is lower than α, the voice data is output to the voice section detection unit 106, and step S
Transition to 303. On the other hand, when the audio level is higher than α, a volume control signal indicating that the volume of the audio signal output from the audio output unit 113 is reduced is output to the volume control unit 111, and the process proceeds to step S311.

In step S303, the voice section detection unit 1
06 detects a voice section which is considered to be uttered by the speaker from the voice data output from the voice input level detection unit 104. If the voice section is shorter than t, the voice section detection unit 106 outputs the voice data to the word recognition unit 107,
The process proceeds to step S304. On the other hand, when the voice section is longer than t, the voice output unit 113 is sent to the volume control unit 111.
A volume control signal indicating that the volume of the audio signal output from is reduced is output, and the process proceeds to step S311.

In step S304, the word recognition unit 107
However, the voice data output from the voice section detection unit 106 is subjected to voice recognition processing using a recognition dictionary. For example, if the electronic device is a television, the recognition dictionary may include, as an example, voice commands that represent the operation content of the television such as “XX channel”, “stereo output”, and “input switching” as standard voice data. It is recorded and word recognition part 1
Reference numeral 07 compares the standard voice data with the voice data output from the voice section detection unit 106 to obtain standard voice data closest to the voice data of the speaker, and determines the reliability of the word and the word as a recognition result. Output to the unit 108.

In step S305, the recognition result judging unit 1
08 compares the reliability of the recognized word output from the word recognition unit 107 with φ, and if the reliability is greater than φ, it is determined that the recognition is successful, and the process proceeds to step S306.
On the other hand, when the reliability is smaller than φ, it is determined that the recognition has failed, the recognized word is rejected, and the volume control signal indicating that the volume of the voice signal output from the voice output unit 113 to the volume control unit 111 is reduced. Is output and step S311
Transition to.

In step S306, the recognition result judging unit 1
08, analyzes the content of the command of the recognition word, and when the content is a command for controlling the video output of the electronic device, outputs a video output control signal to the video output control unit 114 and transitions to step S307. On the other hand, when the content is a command for controlling the audio output of the electronic device, the audio output control signal is output to the audio output control unit 109, and the process proceeds to step S309.

In step S307, the video output controller 1
When the video output control signal is input from the recognition result determination unit 108, the switching unit 14 switches the channel of the video source output from the video output unit 115 according to the content.

In step S308, the video output section 115
Displays the video source switched by the video output control unit 114 on the screen and returns to the voice input waiting state S300.

In step S309, when the voice output control signal is input from the recognition result determination unit 108, the channel switching unit 110 outputs the voice output unit 113 according to the content of the voice output control signal.
Switch the channel of the audio source output from.

In step S310, the voice output unit 113
Outputs the audio source switched by the channel switching unit 110, and returns to the audio input waiting state S300.

In step S311, the volume control unit 111
However, when the volume control signal is input from the voice input level detection unit 104, the voice section detection unit 106, or the recognition result determination unit 108, the volume of the voice signal output from the voice output unit 113 is reduced, and the voice input waiting is performed. Return to state S300.

As described above, according to the embodiment of the present invention, when a voice is output and an electronic device having a voice recognition function is operated by voice, the voice input level detection unit 104 outputs the voice in step S302. When it is determined that the sound level is too high, the sound level detection unit 104 outputs a sound volume control signal to the sound volume control unit 111, which indicates that the sound signal output from the sound output unit 113 is to be decreased in sound volume. The volume control unit 111 includes a voice input level detection unit 104.
When the volume control signal is input from, the volume of the audio signal output from the audio output unit 113 is reduced. That is, it is possible to quickly control the volume of the audio signal output from the electronic device body without performing the processing of step S303 and subsequent steps, and it is possible to improve the operability of the electronic device.

In step S303, when the voice section detection unit 106 determines that the voice section is too long, the voice section detection unit 106 outputs the voice signal output from the voice output unit 113 to the volume control unit 111. A volume control signal that indicates to reduce the volume is output. Volume control unit 11
When the volume control signal is input from the voice section detection unit 106, 1 reduces the volume of the voice signal output from the voice output unit 113. That is, it is possible to quickly control the volume of the audio signal output from the electronic device body without performing the processing of step S304 and subsequent steps, and it is possible to improve the operability of the electronic device.

As described above, since the volume of the electronic device is lowered without performing the processing of step 303 or step 304 or higher, the operability of the electronic device can be speeded up.

Also, when the power of the electronic device is turned on or at step S
When the audio source output from the audio output unit 113 is switched in 310, the parameter change notification unit 11
2 outputs a parameter change notification signal based on the set value of the volume set by the volume control unit 111. The voice input level detection unit 104 appropriately changes the level parameter based on the parameter change notification signal. The voice section detection unit 106 also appropriately changes the section detection parameter based on the parameter change notification signal.

As a result, even if the volume of the voice signal output from the electronic device main body changes and the volume of the voice signal input to the voice input unit 101 changes, the voice signal clearly unsuitable for the recognition process is generated. The detection can be done accurately,
Since the time-consuming recognition process does not have to be performed, the operability of the device can be improved.

The present invention is a program for causing a computer to execute the functions of all or part of means (or devices, elements, circuits, units, etc.) of the electronic apparatus of the present invention described above. It is a program that works in cooperation with.

Further, the present invention is a medium carrying a program for causing a computer to execute all or some of the functions of all or some of the means of the electronic apparatus of the present invention described above, which is readable by the computer and ,
The read program is a medium that executes the function in cooperation with the computer.

Some means (or devices, elements, circuits, parts, etc.) of the present invention means some of the plurality of means, or one of the means. It means a part of the functions.

Further, some devices (or elements, circuits, parts, etc.) of the present invention mean some devices out of the plurality of devices, or one device. , Means a part of means (or an element, a circuit, a part, etc.) or means a part of the function of one means.

A computer-readable recording medium in which the program of the present invention is recorded is also included in the present invention.

Further, one usage form of the program of the present invention may be a mode in which the program is recorded in a computer-readable recording medium and operates in cooperation with the computer.

Further, one usage form of the program of the present invention may be a mode in which the program is transmitted through a transmission medium, read by a computer, and operates in cooperation with the computer.

Further, the data structure of the present invention includes a database, a data format, a data table, a data list, a data type and the like.

Further, the recording medium includes a ROM and the like, the transmission medium includes the transmission medium such as the Internet,
Light, radio waves, sound waves, etc. are included.

Further, the computer of the present invention described above is
The hardware is not limited to pure hardware such as a CPU, and may include firmware, an OS, and peripheral devices.

As described above, the configuration of the present invention may be realized by software or hardware.

[0076]

As described above, according to the present invention, the following effects can be obtained.

First, when the voice level is higher than a predetermined value, the volume of the voice signal output from the voice output means is reduced, so that the subsequent recognition processing is not performed and the output voice is output. The volume can be reduced.
That is, when the sound level is higher than a predetermined value, the output sound can be quickly controlled.

Secondly, when the voice section is longer than a predetermined value, the volume of the voice signal output from the voice output means is reduced, so that the subsequent recognition processing is not performed and the output voice is output. The volume can be reduced. That is, when the voice section is larger than a predetermined value, the output voice can be quickly controlled.

Thirdly, when the volume of the output voice is changed, one or both of the level parameter and the section detection parameter are appropriately changed based on the parameter change notification signal output from the parameter change notifying means. Therefore, even if the volume of the output voice is changed, the voice section and the voice input level can be accurately detected. This allows
Even if the volume of the output voice is changed, it is possible to accurately detect the voice that is obviously inappropriate for the recognition process, and to quickly control the output voice.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an embodiment of the present invention.

FIG. 2 is a flowchart showing a processing procedure according to the embodiment of the present invention.

FIG. 3 is a block diagram showing an electronic device of a conventional example of the present invention.

[Explanation of symbols]

101 voice input unit 102 voice detector 103 microphone 104 voice input level detector 105 voice recognition unit 106 voice section detector 107 word recognition unit 108 Recognition result determination unit 109 Audio output control unit 110 channel switching unit 111 Volume control unit 112 Parameter change notification section 113 voice output section 114 video output controller 115 Video output section

Continuation of the front page (51) Int.Cl. ⁷ Identification code FI theme code (reference) G10L 15/10 15/28 (72) Inventor Fuji Saki ▼ Katsumi Katsumi Osaka Prefecture 1006 Kadoma, Kadoma Matsushita Electric Industrial Co., Ltd. Inner F term (reference) 5D015 AA04 BB01 DD02 KK01 LL12

Claims (10)

[Claims] 1. An electronic device equipped with voice recognition means, voice output means for outputting voice, and volume control means for controlling the volume of the voice output means detects the start of a voice command of a speaker. Voice detection means,
A voice collection unit that collects voice, a voice input level detection unit that detects the level of the collected voice, and controls the volume control unit based on the level, and a recognition result obtained from the voice recognition unit. An electronic device including at least a recognition result determining means for outputting a voice output control signal for executing the voice command of the speaker based on the above. 2. The voice input level detecting means holds a predetermined level parameter so as to correspond to a voice command level of an input voice, and if the voice level is higher than the level parameter, The electronic device according to claim 1, wherein a volume control signal for reducing the volume of voice is output to the volume control means. 3. An electronic device equipped with voice recognition means, voice output means for outputting a voice, and volume control means for controlling the volume of the voice output means detects the start of a voice command of a speaker. Voice detection means,
A voice collecting unit that collects voice, a voice section detecting unit that detects a voice section of the collected voice and controls the volume control unit based on the length of the voice section, and a voice section that are obtained from the voice recognizing unit. Based on the recognition results
An electronic device comprising: a recognition result determining unit that outputs a voice output control signal for executing the voice command of the speaker. 4. The voice section detecting means holds a predetermined section detection parameter so as to correspond to a voice section of a voice command of an input voice, and when the voice section is longer than the section detection. The electronic device according to claim 3, wherein a volume control signal for reducing the volume of voice is output to the volume control means. 5. The voice detecting means starts a voice command from a sensor for detecting the presence of a person in a predetermined area based on the voice collecting means, or the speaker itself by a method other than voice. The electronic device according to claim 1 or 3, which is configured by a switch or a sensor capable of notifying the user of the fact. 6. A volume setting value detecting means for detecting a setting value of the volume output by the device, and the level parameter or the section detecting parameter is changed based on the setting value detected by the volume setting value detecting means. The electronic device according to claim 1, further comprising a parameter changing unit that indicates that the electronic device is operated. 7. The electronic device according to claim 1, wherein a voice detecting means for detecting the start of a voice command of a speaker, a level of collected voice are detected, and the volume control means is controlled based on the level. All or one of the voice input level detecting means for controlling and the recognition result judging means for outputting at least a voice output control signal for executing the voice command of the speaker based on the recognition result obtained from the voice recognizing means. A program that causes a computer to function as a department. 8. The electronic device according to claim 3, wherein the voice detection means for detecting the start of a voice command of the speaker, the voice section of the collected voice are detected, and based on the length of the voice section, A voice section detecting means for controlling the volume control means,
A program for causing a computer to function at least as a whole or part of a recognition result determination unit that outputs a voice output control signal for executing a voice command of the speaker based on the recognition result obtained from the voice recognition unit. . 9. The electronic device according to claim 1, wherein a voice detecting means for detecting a start of a voice command of a speaker, a level of collected voice are detected, and the volume control means is controlled based on the level. All or one of the voice input level detecting means for controlling and the recognition result judging means for outputting at least a voice output control signal for executing the voice command of the speaker based on the recognition result obtained from the voice recognizing means. A medium that carries a program for causing a computer to function as a unit, and is characterized by being processable by the computer. 10. The electronic device according to claim 3, wherein the voice detection means for detecting the start of a voice command of the speaker, the voice section of the collected voice are detected, and based on the length of the voice section, A voice section detecting means for controlling the volume control means, and a recognition result judging means for outputting at least a voice output control signal for executing a voice command of the speaker based on a recognition result obtained from the voice recognizing means. A medium carrying a program for causing a computer to function as all or a part of the above, and characterized by being processable by the computer.