JP2000050149A - Video camera - Google Patents

Abstract

PROBLEM TO BE SOLVED: To realize a unit operation by means of sound input without installing a microphone for operation sound input and a speaker for operation confirmation, to miniaturize the constitution and to reduce the cost. SOLUTION: Microphones 1L and 1R for photographing sound input are shared for operation sound input, and operation sound inputted form the microphones 1L and 1R is recognized in DSP 22 of a sound input operation processing block 60. A micro computer 24 generates an operation control signal 25 corresponding to operation sound which DSP 22 recognizes and controls a recording/reproducing device 15 by the control signal 25. DSP 22 generates an operation confirmation signal 26 for informing a user of the recognition result of operation sound and outputs it from a regular reproduction speaker 13.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a video camera having at least a microphone for audio input, such as a camera-integrated video tape recorder.

[0002]

2. Description of the Related Art In recent years, the operation of various electronic devices has become more complicated as they have become more sophisticated and multifunctional, and it has become difficult for anyone to easily operate them. As one solution to this problem, a user utters a voice of a word indicating the operation content of the electronic device (hereinafter, appropriately referred to as an operation voice), and the electronic device recognizes the operation voice and operates. Techniques for improving operability are known.

FIG. 4 shows photographing sound input microphones 101L and 101R used to capture and record normal photographing sound (hereinafter, appropriately referred to as photographing sound), and operation sound for capturing the operation sound. An input microphone (for inputting voice for which voice recognition is performed) 120, and the photographing voice input microphones 101L and 101L.
The photographing sound input from the R is recorded on a recording medium such as a magnetic tape, while the operation sound input from the operation sound input microphone 120 realizes the operation of the device, and the camera-integrated digital video tape recorder ( Hereinafter, a general configuration of a camera-integrated digital VTR will be described.

[0004] Of the blocks surrounded by broken lines in FIG.
The upper block shows a recording / reproducing signal processing block 100 which is a main component of the camera-integrated digital VTR, and the lower block realizes control of the operation of the device by taking in the operation voice from the user (user). Input operation processing block 130 is shown.

In a recording / reproducing signal processing block 100, a video signal from a CCD (solid-state imaging device) 102 is
The digital signal is converted into a digital video signal 104 after undergoing various kinds of signal processing in the camera signal processing circuit 103, and is sent to the recording video / audio signal processing circuit 105.

[0006] The photographing sound signals input from the L (left) and R (right) two-channel photographing sound input microphones 101L and 101R are amplified by corresponding preamplifiers 116L and 116R, respectively, and further analog / digital. The digital signal is converted into a digital audio signal 107 by a converter (ADC) 106, and the recording video / audio signal processing circuit 10
Sent to 5.

The recording system video / audio signal processing circuit 105
Are digital video signal 104 and digital audio signal 1
07, the corresponding video / video compression processing, audio-based shuffling processing, framing processing of video and audio signals to be recorded on a tape, and the like are performed to generate a recorded video / audio signal 108. This recording video / audio signal 108 is sent to the switch SW114.

The switch SW114 at the time of recording has its switched terminal switched to the REC (recording) side, whereby the recorded video / audio signal 108 is switched to the switch SW1.
The data is sent to a built-in recording / reproducing device 115 such as a video tape recorder (VTR) via the.

The recording / reproducing apparatus 115 at the time of recording converts the input recording video / audio signal 108 into an RF signal, and records the RF signal on a magnetic tape by an electro-magnetic conversion system including a rotating drum and a magnetic head. I do.

Conversely, the recording / reproducing apparatus 115 at the time of reproduction reproduces an RF signal recorded on a magnetic tape by an electro-magnetic conversion system, and converts the RF signal into a video / audio signal (reproduced video / video).
The signal is converted to an audio signal 109) and sent to the switch SW114.

The switch SW114 during the reproduction
In this case, the switched terminal is switched to the PB (reproducing) side, whereby the reproduced video / audio signal 109 is sent to the reproducing video / audio signal processing circuit 110.

The reproduction video / audio signal processing circuit 110
It performs signal processing opposite to that of the recording video / audio signal processing circuit 105. The reproduced video / audio signal 109 is subjected to video and audio signal deframing processing, video expansion processing, and audio deshuffling processing. And so on.

The reproduction video / audio signal processing circuit 110
Is output from, for example, an external output terminal or sent to a display processing system (not shown) to be displayed. An audio signal (photographed audio signal) 111 output from the reproduction video / audio signal processing circuit 110 is converted into an analog audio signal by a digital / analog converter (DAC) 112 and amplified by an amplifier 117. Thereafter, the sound is converted into electric sound by the reproduction speaker 113 and output as sound. The reproduction speaker 113 is connected to the recording / reproduction device 11.
Reference numeral 5 denotes a speaker for outputting the photographed sound reproduced from the magnetic tape.

Next, in a voice input operation processing block 130, the voice signal input from the operation voice input microphone 120 is amplified by a preamplifier 128, and further converted into a digital voice signal by an analog / digital converter (ADC) 121. The signal is converted and sent to a DSP (Digital Signal Processor) 122.

The DSP 122 extracts a voice pattern (hereinafter, referred to as an input voice pattern) from the input digital voice signal, and uses the input voice pattern and an operation voice recognition registered in a ROM (read only memory) 123 in advance. Of the voice pattern (hereinafter, referred to as an operation voice pattern) is analyzed. The signal of the pattern matching analysis result is sent to the microcomputer 124.

In the ROM 123, operation voice patterns of a group of words prepared in advance for voice-based device operation are registered. The word group prepared for the operation includes, for example, “Rokuga”, “Tomare”,
"Saisei" and the like are conceivable. Note that “Rokuga” is a word for instructing a device to start recording,
“Tore” is a word for instructing the device to perform a stop operation, and “Saisei” is a word for instructing the device to perform a reproduction start operation.

In the pattern matching analysis by the DSP 122, when the input voice pattern and the operation voice pattern match, the microcomputer 124 generates the operation control signal 125 and sends it to the recording / reproducing device 115, and at the same time, generates the operation confirmation signal 126. It is generated and sent to the operation confirmation speaker 127.

The operation control signal 125 indicates an operation corresponding to the matched operation voice pattern in the recording / reproducing apparatus 1.
15 is a control signal for causing
The operation confirmation signal 126 is input to the operation voice input microphone 1
This is a voice signal for notifying the user that the device has recognized the content of the voice input from the device 20 (recognized as an operation voice). The operation confirmation signal 126 is a signal such as a buzzer sound or a so-called beep sound. Therefore, the operation confirmation speaker 127 is a speaker that outputs these buzzer sounds or beep sounds.

On the other hand, in the pattern matching analysis in the DSP 122, if the input voice pattern and the operation voice pattern do not match, the DSP 122 and the microcomputer 124 maintain the current state and continue the pattern matching analysis.

[0020]

As shown in FIG.
Conventionally, an operation voice input microphone for capturing operation voices (voices recognized by voice) and a shooting voice input microphone for capturing shooting voices (original voices to be recorded on a recording medium) are separately prepared. It is common to do.

As described above, the operation voice input microphone and the photographing voice input microphone are separately prepared because the operation voice is sent to a DSP or the like to perform voice recognition and judgment. This is because the sound is subjected to signal processing for recording on a recording medium such as a magnetic tape by the recording / reproducing apparatus. In addition, for example, in order to prevent the operation sound from being recorded on the recording medium together with the photographing sound during the actual shooting, and to make it possible to reliably execute the device operation instruction by the operation sound, The input microphone and the photographing voice input microphone are separately prepared, and the operation voice input microphone and the photographing voice input microphone are arranged separately from each other.

However, in practice, it is difficult to completely isolate both microphones in a small device such as a digital VTR with a built-in camera, and an operation voice (voice to be recognized) is also recorded at the same time as a shooting voice. There is a drawback that it often happens. Also, from the viewpoint of miniaturization and cost reduction, it is disadvantageous to mount two types of microphones separately.

Further, when an operation function using voice is provided in a digital VTR with a built-in camera, it is necessary to inform the user whether or not the operation voice has been correctly recognized as a result of voice recognition. However, an operation confirmation speaker for that purpose is separately required. However, in recent years, a camera-integrated digital VTR and the like generally include a built-in speaker for reproducing photographed sound, and providing a speaker for operation confirmation in addition to the speaker for reproduction is downsized and cost-effective. It is not desirable from the viewpoint of reduction.

Therefore, the present invention has been made in view of such a situation.
To provide a video camera such as a TR that realizes device operation by voice input without providing a microphone for operation voice input, a speaker for operation confirmation, and the like, and at the same time, realizes miniaturization and cost reduction of the configuration. Aim.

[0025]

SUMMARY OF THE INVENTION A video camera according to the present invention is a video camera provided with a microphone capable of inputting at least a sound at the time of photographing. The video camera recognizes an audio signal input from the microphone and obtains a recognition result. The above-mentioned problem is solved by controlling each unit based on the above.

In the video camera of the present invention, when a voice signal input from a microphone is recognized by voice recognition means, a predetermined voice is output from at least a speaker capable of reproducing voice recorded at the time of shooting. By
The above-mentioned problem is solved.

[0027]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described with reference to the drawings.

As one embodiment to which the video camera of the present invention is applied, a camera-integrated digital VTR will be described. Hereinafter, the camera-integrated digital VTR of this embodiment will be described.

FIG. 1 shows an L-channel microphone 1
This embodiment includes L and R channel side microphones 1R, and the microphones 1L and 1R capture actual photographing voices and also capture operation voices for realizing operation of devices by voices. 1 shows the overall configuration of a camera-integrated digital VTR.

Of the blocks surrounded by broken lines in FIG.
A block 50 is a signal processing block (hereinafter referred to as a camera-integrated VTR) of the entire camera-integrated digital VTR of the present embodiment.
TR signal processing block 50), and block 6
0 is a processing block for realizing the operation of the device by voice input (hereinafter referred to as a voice input operation processing block 60).
Is shown.

First, the camera-integrated VTR, which is a main component of the camera-integrated digital VTR of the present embodiment, is described.
The TR signal processing block 50 will be described.

Camera-integrated VTR signal processing block 50
, A video signal from a CCD (solid-state imaging device) 2 undergoes various kinds of signal processing in a camera signal processing circuit 3, is converted into a digital video signal 4, and is sent to a recording video / audio signal processing circuit 5.

The audio signals input from the L (left) and R (right) two-channel microphones 1L and 1R are amplified by the corresponding preamplifiers 16L and 16R, respectively, and further converted to an analog / digital converter (ADC). After being converted into a digital audio signal 7 in step 6, it is sent to the recording video / audio signal processing circuit 5 and also to an audio input operation processing block 60 to be described later.

The recording video / audio signal processing circuit 5
The digital video signal 4 and the digital audio signal 7 are subjected to the corresponding video video compression processing, audio shuffling processing, framing processing of video and audio signals to be recorded on tape, and the like. An audio signal 8 is generated. This recorded video / audio signal 8 is sent to the switch SW14.

The switch SW14 at the time of this recording is
The switched terminal is switched to the REC (recording) side. As a result, the recorded video / audio signal 8 is switched to the switch SW1
4 to a recording / reproducing device 15 such as a video tape recorder (VTR).

The recording / reproducing device 15 at the time of recording converts the input recorded video / audio signal 8 into an RF signal, and records the RF signal on a magnetic tape by an electro-magnetic conversion system including a rotating drum and a magnetic head. I do.

Conversely, the recording / reproducing device 15 at the time of reproduction reproduces the RF signal recorded on the magnetic tape by the above-mentioned electro-magnetic conversion system, and converts the reproduced RF signal into a video / audio signal (reproduced video / audio). Switch to the audio signal 9) and switch SW14
Send to

The switch SW14 at the time of the reproduction is
The switched terminal is switched to the PB (reproduction) side. As a result, the reproduced video / audio signal 9 is sent to the reproduction video / audio signal processing circuit 10 via the switch SW14.

The reproduction video / audio signal processing circuit 10 performs signal processing reverse to that of the recording video / audio signal processing circuit 5, and performs decoding of video and audio signals on the reproduction video / audio signal 9. It performs framing processing, video-based decompression processing, and audio-based deshuffling processing.

The video signal output from the reproduction video / audio signal processing circuit 10 is output from, for example, an external output terminal or sent to a display processing system (not shown) for display.

Also, the reproduction video / audio signal processing circuit 10
The audio signal (photographed audio signal) output from the
0 through a digital / analog converter (DAC) 12
The digital / analog converter 12 converts the audio signal into an analog audio signal, which is further amplified by an amplifier 17, converted into an electric audio signal by a speaker 13, and output as audio. The adder 30 is provided to add an operation confirmation signal (voice signal) from the DSP 22 of the voice input operation processing block 60 described later to the voice signal sent to the speaker 13.
The audio signal 11 output from the device includes a photographing audio signal and an operation confirmation signal. Thereby, the speaker 13
Is used for both the reproduction of the photographed sound taken out of the magnetic tape by the recording / reproducing device 15 and the reproduction of the operation confirmation sound based on the operation confirmation signal. Since the speaker 13 is a speaker that can reproduce the photographed sound, it can output not only a buzzer sound and a beep sound but also various sounds such as words.

Next, the voice input operation processing block 60 will be described.

The digital audio signal 7 from the digital / analog converter 6 is supplied to an audio input operation processing block 6
0 is also sent to the DSP 22.

The DSP 22 extracts a voice pattern (input voice pattern) from the input digital voice signal 7, and registers the input voice pattern and one of standard voice patterns in a ROM (read only memory) 23 in advance. The matching with the voice pattern for operation voice recognition (operation voice pattern) is analyzed. The signal of the pattern matching analysis result is sent to the microcomputer 24 as a recognition determination output 45.

The ROM 23 stores, as one of the standard voice patterns, an operation voice pattern corresponding to a word prepared in advance for operating the device by voice, and a word prepared in advance for operation confirmation and voice dialog described later. , A confirmation voice pattern, a conversation voice pattern, and the like corresponding to.
In addition, as the words prepared for the above-mentioned device operation,
In Japanese, for example, various word groups generally used to represent the operation of the device, such as “Rokuga”, “Ishi (or Tomari)”, and “Saisei”, can be considered.
Note that “Rokuga” mentioned here as an example is a word for instructing the device to perform a recording start operation, “Ishishi” is a word for instructing the device to perform a stop operation,
“Saisei” is a word for instructing a device to perform a reproduction start operation. In addition, as words prepared for the above-mentioned operation confirmation, for example, “Rokushigashiru”, “Ishigashishi”, “Saiseishi”, etc. can be mentioned.
The words prepared for the voice dialogue include, for example, "Do you want to do something?", "Do you want to do something?", "Do you want to make something?" Of course, these are merely examples, and various other audio patterns are registered in the ROM 23.

In the pattern matching analysis by the DSP 22, when the input voice pattern and the operation voice pattern match, the DSP 22 sends a recognition determination signal 45 indicating the match to the microcomputer 24 and generates the operation confirmation signal 26. And sends it to the adder 30.

Here, the operation confirmation signal 26 is a signal for notifying the user that the device has recognized the content of the voice input from the microphones 1L and 1R (recognized as the operation voice). The operation confirmation signal 26 may be a signal for generating a buzzer sound, a so-called beep sound, or the like. In the present embodiment, the operation confirmation signal 26 corresponds to a word for causing the user to confirm the content recognized as the operation sound. (Hereinafter, referred to as a confirmation voice signal) or a voice signal corresponding to a word for performing a voice dialogue with the user (hereinafter, referred to as a dialogue voice signal). In addition, as confirmation voices, for example, "Rokugashiru", "
You can mention "I'll do it", and the dialogue voices are, for example, "Do you want to do it?" be able to.

As described above, when outputting the confirmation voice signal as the operation confirmation signal 26, the DSP 22
A confirmation sound pattern corresponding to the confirmation sound signal is read from the standard sound patterns registered in M23,
A confirmation voice signal is generated from this confirmation voice pattern and output. For example, when outputting the confirmation sound signal of "Rokugashiru", the DSP 22 reads the confirmation sound pattern of "Rokugashiru" from the ROM 23, and from this confirmation sound pattern, the confirmation sound signal of "Rokugashiru". Is generated and output. When outputting the dialogue voice signal as the operation confirmation signal 26, the DSP 22 reads the dialogue voice pattern corresponding to the dialogue voice signal from the standard voice patterns registered in the ROM 23, and reads out the dialogue voice pattern from the confirmation voice pattern. Generate and output a confirmation voice signal. For example, the DSP 22 that outputs the dialogue voice signal “Rokusugare?” Reads out the dialogue voice pattern “Rokusugare?” From the ROM 23, and reads “Rokugarashita” from this dialogue voice pattern. ?? is generated and output.

In the present embodiment, the example in which the DSP 22 generates the operation confirmation signal is described. However, the microcomputer 24 may generate the operation confirmation signal using the standard voice pattern in the ROM 23. .

On the other hand, in the pattern matching analysis in the DSP 22, the input voice pattern and the operation voice pattern match, and when the microcomputer 24 receives a recognition determination signal indicating the matching from the DSP 22, the microcomputer 24 outputs the operation control signal 25. It is generated and sent to the recording / reproducing device 15.

Here, the operation control signal 25 is a control signal for causing the recording / reproducing apparatus 15 to perform an operation corresponding to the matched operation voice pattern. For example, if the matched operation voice pattern is the voice pattern “Rokuga”, the microcomputer 24
An operation control signal 25 for starting the recording operation is sent to the recording / reproducing device 15, and for example, when the matched operation sound pattern is a “saisei” sound pattern, the microcomputer 24 sends a signal to the recording / reproducing device 15. Control signal 25 for starting playback operation
Send.

As described above, in the present embodiment, by operating the recording / reproducing apparatus 15 based on the operation control signal 25, the operation control of the recording / reproducing apparatus 15 by voice is realized. In the description so far,
Although an example in which one operation is made to correspond to one word as the operation voice is described, a complicated operation in which a plurality of operations are combined is performed on the recording / reproducing device 15 by an operation voice in which a plurality of words and the like are combined. It is also possible to do it. For example, when an operation voice of “5 Fengoku ga shishi” is input, the recording / reproducing device 15 is controlled to stop after recording for 5 minutes. For example, an operation voice of “Saiyo Return Saisei” is input. In such a case, the recording / reproducing device 15 is controlled so as to return to the beginning of the tape and start reproduction.

In the pattern matching analysis in the DSP 22, if the input voice pattern and the operation voice pattern do not match, the DSP 22 and the microcomputer 24
Will maintain the current state and continue the pattern matching analysis.

In the present embodiment, both the actual photographing voice and the operation voice are shared by the common microphones 1L and 1R.
Therefore, there is a possibility that the recording / reproducing apparatus 15 may record not only the photographing voice but also the operation voice. This problem also occurs when the microphone for photographing voice input and the microphone for operation voice input are close to each other in the configuration shown in FIG.

In the present embodiment, in order to solve such a problem, when the DSP 22 detects that the input voice pattern and the operation voice pattern match, the microcomputer 24 operates, for example, a part including the operation voice. Output an operation control signal 25 for controlling the recording / reproducing device 15 so as not to record the
An operation control signal 25 for controlling the recording / reproducing device 15 so as to later overwrite the portion where the operation voice has been recorded is output.

The manner in which the microcomputer 24 controls the recording / reproducing device 15 will be specifically described with reference to FIG.

FIG. 2A shows a case where recording is started in response to a user's voice, for example, "Rokuga" from a state in which the recording / reproducing device 15 is stopped or in a recording pause mode (recording mode). FIG. 7 shows the transition of the operation mode of the recording / reproducing device 15). In FIG. 2A, for example, when the user utters the voice of “Rokuga” at the timing of T1 in the figure, the DSP 22 causes the input voice pattern of the “Rokuga” and the “Rokuga” of the operation voice pattern. Detects a match with The microcomputer 24 that has received the result of the coincidence detection uses the operation control signal 25 to operate the recording / reproducing device 15 at a timing T2 in the figure after a predetermined time has elapsed from the timing T1 or at the timing T2 in FIG. Start recording.
As a result, the voice of “Rokuga” by the user is not recorded in the recording / reproducing device 15.

FIG. 2B shows a case in which the recording is stopped in response to the user's voice, for example, "stop" from the state in which the recording / reproducing apparatus is in the recording mode (stop recording or temporarily stop recording). When moving to stop mode)
5 shows the operation mode transition of the recording / reproducing apparatus 15.
In FIG. 2B, for example, when the user utters the voice of “Ishi” at the timing of T3 in the figure, the DSP
Reference numeral 22 detects the coincidence between the input voice pattern of the "teishi" and the "teishi" of the operation voice pattern. The microcomputer 24 that has received the result of the coincidence detection sends the operation control signal 25 to the recording / reproducing device 15 at a timing T4 in the figure after a lapse of a predetermined time from the timing T3, or at a timing T4 in FIG. After the recording position is returned to the position of T3 in the figure, the recording is stopped or the recording is temporarily stopped. In the recording stop or recording pause mode, if recording is performed later, T in FIG.
It is overwritten from 3 to T4. In other words, in the case of FIG. 2B, the voice of "Ishishi" uttered by the user during the period from T3 to T4 in the figure is recorded, but as described above, from T4 to T3 in the figure. Is returned and overwritten thereafter, so that the sound of the user is not recorded in the recording / reproducing device 15 (is not recorded).

As described above, in the camera-integrated digital VTR according to the present embodiment, the operation mode of the recording / reproducing device 15 is controlled in accordance with the content of the operation sound.
It is possible to easily prevent the operation voice from being recorded.

On the other hand, since the operation voice from the user is not recorded at the time of reproduction, the operation voice is transmitted to the recording / reproducing device 15 when the match between the input voice pattern and the operation voice pattern is detected. An operation corresponding to the pattern may be performed.

It is to be noted that the input voice from the user is transmitted to the DSP 22
Is recognized as an operation voice, a confirmation voice or a dialogue voice corresponding to the confirmation voice signal 26 is output from the speaker 13 as described above. Therefore, for example, recording is actually started from the above T2, or The recording is stopped after the user has confirmed that the device has correctly recognized the operation by the confirmation voice or the dialogue voice. That is, the microcomputer 2
The user 4 actually controls the operation of the recording / reproducing device 15 by the operation control signal 25 after the user confirms that the device has correctly recognized the operation by the confirmation voice or the dialogue voice. As described above, as a method for the microcomputer 24 to know that the confirmation has been made by the user, for example, after a confirmation sound is output from the speaker 13, a new operation sound is input from the user for a predetermined time or an operation sound is input again When there is no, there is a method in which the confirmation is made by the user or, for example, after the confirmation voice is output from the speaker 13, for example, a voice input of “OK” or “OK” is made by the user. Sometimes, if the above user confirms,
Further, for example, a method may be considered in which the confirmation is made by the user when a confirmation sound is output from the speaker 13 and then the user inputs an input from a predetermined confirmation button or confirmation switch (not shown). .
Note that these methods are merely examples, and various other methods are conceivable.

For example, if a voice uttered by another person other than the user contains the same word as the operation voice, the voice of the other person is erroneously detected as the operation voice of the user, and There is a possibility that an operation not intended by the user may be performed. For this reason, in the present embodiment,
By registering a pattern for identifying a speaker (identifying a user's voice) in the ROM 23, erroneous recognition is avoided. In other words, a rewritable memory that does not require a memory holding operation is used as the ROM 23 in FIG. 1, and an operation voice pattern learned by the user by inputting the operation voice is registered in advance. This can prevent erroneous recognition of the voice of another person as the operation voice of the user.

Further, for example, a sound to be recorded as a normal sound, that is, “Rokuga”, “Ishishi”, “Saisei” or the like is uttered by a user as a photographed sound, and the sound is captured by a microphone. In such a case, if these photographed sounds are erroneously detected as operation sounds, an operation not intended by the user is performed, and the photographed sounds are not recorded despite the desire to record them. According to advances in speech recognition technology in
Such misperceptions are on the decline. Further, for example, a switch 31 is provided as shown in FIG.
It is also possible to consider a method in which a voice operation is accepted only when 1 is turned on. Note that this method is an example, and various other methods are conceivable.

Next, FIG. 3 shows a specific configuration example of the DSP 22 and the ROM 23 in the voice input operation processing block 60 shown in FIG.

In FIG. 3, digital audio signal 7 from analog / digital converter 6 in FIG.
22. Here, in the present embodiment, both the actual shooting sound and the operation sound are shared by the common microphones 1L and 1L.
The digital audio signal 7 input to the DSP 22 is either a photographed audio signal only, an operation audio signal only, or a mixed signal of the photographed audio signal and the operation audio signal. Therefore,
In the DSP 22, the supplied digital audio signal 7
Only the operation voice signal must be extracted from.

Therefore, the digital audio signal input to the DSP 22 is first input to the pre-processing unit 41. The pre-processing unit 41 is configured by a filter for extracting only a signal in a voice section from the input digital voice signal 7 and removing noise other than the voice signal. The signal of the voice section extracted by the preprocessing unit 41 is sent to the spectrum analysis / feature parameter extraction unit 42.

The spectrum analysis / characteristic parameter extraction unit 42 performs a spectrum analysis of the signal of the voice section, and uses the obtained spectral components to obtain characteristic patterns necessary for pattern matching in the pattern matching unit 43 in the subsequent stage. And parameters such as pitch waveforms. The feature parameters extracted by the spectrum analysis / feature parameter extraction unit 42 are sent to the pattern matching unit 43 as the input voice pattern.

In the pattern matching section 43, the pattern obtained from the characteristic parameter, that is, the pattern of the voice input from the microphone (input voice pattern) and the operation voice pattern previously stored in the standard voice pattern storage area 44 of the ROM 23 are stored. And it is determined whether or not the patterns match.
The result of the pattern matching by the pattern matching unit 43 is sent to the microcomputer 24 of FIG.

When the input voice pattern matches the operation voice pattern, a signal corresponding to the matched operation voice pattern is sent from the pattern matching unit 43 to the operation confirmation signal generation unit 46. In the operation confirmation signal generator 46, the standard voice pattern storage 4
4, a pattern (confirmation voice pattern) for generating a confirmation voice to be output from the speaker 13 or a pattern (conversation voice pattern) for generating a dialogue voice is extracted from the corresponding operation voice pattern. An audio signal as the operation confirmation signal is generated using these audio patterns. In other words, as a result of the pattern matching, for example, when the input voice pattern “Rokuga” matches the operation voice pattern “Rokuga”, the confirmation voice pattern “Rokugashiru” voice pattern, or The voice pattern of the dialogue voice pattern "Do you want to do?" Is read from the standard voice pattern storage unit 44 of the ROM 23, and the voice signal is output.

As described above, in the embodiment of the present invention, microphones for inputting operation voice and inputting voice for shooting are not separately provided as in the configuration of FIG. By sharing the microphone for operation voice input, not only the number of microphones can be reduced, but also hardware such as preamplifiers and analog / digital converters for operation voice input can be reduced. The cost is reduced and the size of the configuration is also reduced. According to the present embodiment,
Even if crosstalk occurs between the photographing voice signal and the operation voice signal by sharing the microphone, the operation voice can be accurately recognized, and therefore, a reliable voice operation can be performed. It is possible that the operation voice is not recorded.

Further, in the embodiment of the present invention, FIG.
By not using separate speakers for operation confirmation and playback as in the above configuration, the speakers for playback are shared for operation confirmation, thereby reducing the number of speakers and other parts attached to the speakers. As a result, the overall cost is reduced and the size of the configuration is reduced. Further, according to the present embodiment, since the reproduction speaker is used for operation confirmation, not only a conventional buzzer sound and beep sound but also a sound as a word can be used as the operation confirmation sound. Therefore, it is possible to perform confirmation by the voice or confirmation by dialogue with the user.

The present invention is not limited to the camera-integrated digital VTR described as an example in the above embodiment, but includes a microphone for voice recording and a speaker for voice reproduction, and records the voice fetched by the microphone. The present invention can be applied to various electronic devices, for example, a digital still camera capable of recording sound at the time of shooting.

What the microcomputer 24 controls based on the voice input is not limited to the recording / reproducing device 15, but may be another part of the camera-integrated digital VTR. For example, when a display means (for example, a liquid crystal display device or the like) for displaying a photographed image is provided, it is possible to control the display operation of the display means and to control focusing and zooming of a lens system. It is.

[0074]

As is apparent from the above description, according to the present invention, a voice signal input from a microphone capable of inputting a voice at the time of shooting is recognized, and each unit is controlled based on the recognition result. Accordingly, the device operation by voice input is realized without providing a separate microphone for operation voice input, and the configuration is reduced in size and cost is reduced.

Further, according to the present invention, when the voice recognition means recognizes the voice signal input from the microphone, a predetermined voice is output from a speaker capable of reproducing the voice recorded at the time of shooting, so that the voice is separately output. The device operation by voice input is realized without providing an operation confirmation speaker, and the configuration is reduced in size and cost is reduced.

[Brief description of the drawings]

FIG. 1 is a digital V-camera integrated with a camera according to an embodiment of the present invention.
FIG. 2 is a block circuit diagram illustrating a schematic configuration of a TR.

FIG. 2 is a diagram used to describe a state of control of a recording / reproducing device by a microcomputer.

FIG. 3 is a block circuit diagram showing a specific configuration example of a DSP and a ROM in the voice input operation processing block of FIG. 1;

FIG. 4 is a camera-integrated digital VT capable of voice operation.
FIG. 2 is a block circuit diagram illustrating a schematic configuration of R.

[Explanation of symbols]

1L, 1R Microphone for photographing and operation voice input,
2 CCD, 3 camera system signal processing circuit, 5 recording system video / audio signal processing circuit, 6 analog / digital converter, 10 reproduction system video / audio signal processing circuit,
12 digital / analog converter, 13 speaker for playback and operation confirmation, SW14 switch, 15
Recording / reproducing device, 16L, 16R preamplifier, 22
DSP, 23 ROM, 24 microcomputer, 30 adder, 31 switch, 41 preprocessing unit, 42 spectrum analysis / feature parameter extraction unit, 43 pattern matching unit, 44 standard voice pattern storage unit, 46 operation confirmation signal utterance unit, 50
Camera integrated VTR signal processing block, 60 voice input operation block

Claims (6)

[Claims] 1. A video camera comprising at least a microphone capable of inputting a sound at the time of photographing, a voice recognizing means for recognizing a voice signal input from the microphone, and each unit based on a recognition result of the voice signal. And a control means for controlling the video camera. 2. The control means controls at least a recording device, and records other audio signals of the audio signals input from the microphone except for the audio signal recognized by the audio recognition means in the recording device. The video camera according to claim 1, wherein the video camera is operated. 3. The voice recognition means detects a match by comparing a plurality of voice patterns registered in advance with a pattern of a voice signal input from the microphone, and detects a match by the voice recognition means. 2. The video camera according to claim 1, wherein said control means performs control corresponding to the matched audio pattern. 4. The voice recognition means includes: voice section extraction means for extracting a voice section from a voice signal input from the microphone; spectrum analysis means for performing a spectrum analysis of the extracted voice section; A feature parameter extracting unit that extracts a feature parameter of a voice as a pattern of a voice signal input from the microphone; a voice pattern storing unit that stores a plurality of voice patterns; and a pattern of the feature parameter and the voice pattern storing unit. 4. The video camera according to claim 3, further comprising: a pattern comparison / match detection unit that detects a match by comparing with a stored voice pattern. 5. A speaker capable of reproducing at least a sound recorded at the time of photographing, and a predetermined sound generation for outputting a predetermined sound from the speaker when the sound recognition means recognizes a sound signal input from the microphone. 2. The video camera according to claim 1, further comprising: 6. The predetermined voice generating means includes: a confirmation voice for allowing a user to confirm the recognition result by the voice recognition means; or a predetermined voice between the user according to the recognition result by the voice recognition means. Dialogue voice for dialogue,
6. The video camera according to claim 5, wherein the predetermined sound is generated from the speaker.