JP7216621B2 - Electronic devices, programs and speech recognition methods - Google Patents

Description

TECHNICAL FIELD Embodiments of the present invention relate to an electronic device, a program, and a speech recognition method.

2. Description of the Related Art In recent years, there is a growing need for a service for operating devices and searching for information and content by voice (speech-interactive content search service using AI). This search service is rapidly spreading due to the convenience of being able to operate and search for information simply by speaking to a device without holding a remote controller (hereinafter referred to as "remote controller").

Since the devices that can be operated are not limited to devices that the speaker talks to, but all devices in the home, it is expected that the number of business operators and device manufacturers that provide such search services will increase in the future. be.

On the other hand, when giving an instruction to a device equipped with a device capable of displaying information such as a television set (hereinafter referred to as "TV") or a personal computer (hereinafter referred to as "PC") from a position away from the device. , remote control operation is the basis, but it is conceivable to utilize the remote control as sound collecting means when searching for contents, inputting characters, or the like.

As a specific example of utilizing the sound collected by the TV, for example, a microphone is built into the remote control, and the voice uttered by the speaker is collected by the microphone, transmitted from the remote control to the TV main body by wireless communication, and processed (speech recognition ), or that a microphone is built into the TV body so that the TV body directly collects and processes the user's voice.

In the former example, in which a microphone is built into the remote control, the distance between the microphone and the speaker is close, so high-quality voice can be collected and the collected voice can be recognized and processed with high accuracy, while the speaker holds the remote control by hand. The downside is that it is necessary.

In the latter example, in which a microphone is built into the TV main body, the speaker can speak without holding the remote control, but the distance between the microphone and the speaker is large, so the voice collected by the microphone is difficult to control. I don't expect high quality.

Therefore, it is conceivable to provide microphones on both the remote controller and the TV main body, which is good for both.

JP 2006-319797 A

However, if both the remote control (external terminal) and the TV main unit (electronic device) are equipped with microphones (sound collectors), there are cases where the sounds collected by each microphone are input to the TV main unit at the same time (sound collision). There is a problem that it occurs and the collected voice cannot be used well.

For example, when the speaker has a remote control, it is better to use the sound collected by the microphone of the remote control, and when the speaker does not have the remote control, it is better to use the sound collected by the microphone on the TV side. is good. In this way, it is necessary to use different microphones depending on the situation of the speaker.

The problem to be solved by the present invention is to provide sound collectors in both the external terminal and the electronic device to improve the operability of the speaker's instruction, and use a plurality of sound collectors according to the situation of the speaker. To provide an electronic device, a program, and a speech recognition method capable of utilizing speech collected by a sound part.

An electronic device according to an embodiment is an electronic device wirelessly or wiredly connected to an external terminal having a first sound collecting unit that collects a first sound around the device, and comprises a first sound acquiring unit and a second sound collecting unit. section, a second speech acquisition section, a speech recognition section, and a control section. The first sound acquisition unit acquires the first sound collected by the first sound collection unit of the external terminal from the external terminal. The second sound collector collects a second sound around itself. The second sound acquisition unit acquires the second sound collected by the second sound collection unit. The speech recognition unit performs speech recognition processing on input speech. The control unit inputs the voice that matches the preset condition among the first voice and the second voice to the voice recognition unit and causes the voice recognition unit to perform voice recognition processing.

It is a figure which shows the structure of the recording/reproducing apparatus of embodiment. 4 is a flowchart showing a first operation example of the recording/reproducing device; 9 is a flowchart showing a second operation example of the recording/reproducing device; 10 is a flowchart showing a third operation example of the recording/reproducing device;

Hereinafter, embodiments will be described in detail with reference to the drawings.
FIG. 1 is a diagram showing an example of a schematic configuration of a recording/reproducing apparatus 1 according to one embodiment of an electronic device. In this embodiment, the recording/reproducing apparatus 1 including the image display section 14 will be described, but the image display section 14 is not an essential component. When the electronic device is, for example, a digital recorder or a main body of a computer, the electronic device does not include the image display unit 14, and outputs display information to an external image display unit (display) via various cables. do. In addition, the electronic device may be, for example, an air conditioner, a refrigerator, or the like.

The configuration of the recording/reproducing apparatus 1 will be described with reference to FIG. As shown in FIG. 1, the recording/reproducing apparatus 1 is an electronic device that is wirelessly connected to a remote controller 20 (hereinafter referred to as "remote controller 20") as an external terminal, and provides a voice content search service on the network. The main unit 100 of the recording/reproducing apparatus is connected to one or more service servers (servers 200, 201, etc.) as one or more computers to perform the recording/reproducing apparatus via the network NTW. The recording/reproducing device 1 may be connected to the remote controller 20 by wire.

The recording/reproducing apparatus main body 100 is connected to the remote controller 20 by wireless communication such as Bluetooth (registered trademark) and infrared communication. The remote controller 20 may be an information terminal such as a smart phone or a tablet, or a unit having a microphone and a communication function, in addition to being dedicated to the recording/reproducing apparatus 1 as in this example.

The remote controller 20 includes a plurality of buttons 21 for operating the functions of the recording/reproducing apparatus body 100, a signal processing section 22, an IR transmission section 23 as a first transmission section, a microphone 24 as a first sound collection section, and an audio processing section. 25, a Bluetooth communication unit 26 (hereinafter referred to as "BT communication unit 26") as a second transmission unit, and the like. One of the plurality of buttons 21 includes a setting button 21a for calling a setting function and a voice button 21b for activating a voice function.

The signal processing unit 22 generates signals according to pressing of the plurality of buttons 21 . The IR transmission section 23 outputs a signal generated by the signal processing section 22 according to the operation of the voice button 21b by infrared communication. When the voice button 21b is pressed, the signal processing unit 22 outputs a signal for causing the voice function of the recording/reproducing apparatus main body 100 to start a recording operation, that is, an instruction signal for instructing the recording/reproducing apparatus main body 100 to start recording (specific trigger signal).

The microphone 24 has a narrow sound collection range (directivity of about 90°, sound collection distance of about several tens of centimeters), and is activated by the operation of the voice button 21b. Since one voice (mainly voice spoken by the speaker into the microphone 24) is collected, relatively high-quality voice can be obtained.

The audio processing unit 25 digitizes the analog audio collected by the microphone 24 and transfers it to the BT communication unit 26 . The BT communication unit 26 transmits the voice digitized by the voice processing unit 25 to Bluetooth communication. That is, the BT communication unit 26 and the audio processing unit 25 transmit the audio collected by the microphone 24 to the recording/reproducing apparatus body 100 .

The recording/reproducing apparatus main body 100 includes an antenna 50 for receiving terrestrial digital broadcasting, a tuner 51, an OFDM demodulator 52, a signal processing unit 53, a graphic processing unit 58, an audio processing unit 59, an OSD signal generation unit 61, a video display unit 14, Speaker 15, operation unit 16, various terminals not shown (video output terminal, audio output terminal, etc.), various interfaces (IR receiving unit 18, BT communication unit 19, communication interface 73 with LAN and external network NTW (hereinafter referred to as "communication I /F73”)), a main body microphone 81, a control module 65, a hard disk drive 101 (hereinafter referred to as “HDD 101”), etc. The HDD 101 provided inside the device is also called an internal HDD.

The antenna 50 supplies the received digital terrestrial television broadcast signal to a tuner 51 for digital terrestrial broadcasting. The tuner 51 selects a broadcast signal of a designated channel from the supplied broadcast signals and supplies it to an OFDM (orthogonal frequency division multiplexing) demodulator 52 .

The OFDM demodulator 52 demodulates the input broadcast signal of the channel into a digital video signal and audio signal, and then outputs the digital video signal and audio signal to the signal processing unit 53 .

The signal processing unit 53 performs predetermined digital signal processing on the digital video signal and audio signal input from the OFDM demodulator 52 and outputs the processed signal to the graphic processing unit 58 and the audio processing unit 59 .

The graphics processing unit 58 superimposes the OSD (on screen display) signal generated by the OSD (on screen display) signal generating unit 61 on the digital video signal supplied from the signal processing unit 53 , and outputs the superimposed signal to the video processing unit 62 . The graphic processing section 58 selectively outputs the output video signal of the signal processing section 53 and the output OSD signal of the OSD signal generating section 61, or outputs both outputs in combination.

The video processing unit 62 processes the digital video signal input from the graphics processing unit 58 for brightness, luminance, saturation, etc., and outputs the video signal to the video display unit 14 and a video output terminal (not shown). supply. The video processing unit 62 functions as an output unit that outputs video of content to a screen.

The image display unit 14 is, for example, a display or a display panel, and displays an image based on the image signal on the display panel. When an external device is connected to the video output terminal, the video signal supplied to the video output terminal is output to the external device.

The audio processing unit 59 converts the input digital audio signal into an analog audio signal that can be reproduced by the speaker 15, and then outputs the analog audio signal to the speaker 15 to output audio. The analog audio signal is output to the outside via an audio output terminal (not shown) such as a headphone terminal.

The operation unit 16 includes buttons and switches provided on the main body 100 of the recording/reproducing apparatus, and can operate each function of the main body 100 of the recording/reproducing apparatus in substantially the same manner as the remote controller 20 .

More specifically, the operation unit 16 can be directly operated by the user, for example, displaying an EPG (electronic program guide) for viewing a program, recording reservation, and selecting a television broadcast (program) channel (broadcast station) from the EPG (electronic program guide). , Start recording a program (REC), Display a list of programs to play back recorded programs (past program guide), Select to play back recorded programs from the past program guide (up, down, left, and right directions) and a control command corresponding to playback start (PLAY) is input to the control module 65 .

The main body microphone 81 collects a second sound (speaker's voice) around itself (the main body microphone 81) (a range of several meters with directivity at a certain angle in front of the screen of the video display unit 14). It is a sound collecting unit, and collects sound in a wider sound collecting range than the microphone 24 of the remote controller 20 (directivity of about 120°, sound collecting distance of about several meters).

The input sound processing unit 64 digitizes the analog sound collected by the main body microphone 81 and outputs the digitized sound to the control module 65 . The input sound processing unit 64 functions as a second sound acquisition unit that acquires the second sound collected by the main body microphone 81 .

Normally, when the recording/reproducing apparatus main body 100 is in operation, the main body microphone 81 always collects sound in a sound-collectable state (active state), and when the voice button 21b of the remote controller 20 is pressed, it is in an inactive state (sound-collecting operation is activated). (stopped state), the microphone 24 of the remote controller 20 is activated, and the voice (first voice) collected by the microphone 24 is obtained from the remote controller 20 .

In addition, even when the voice button 21b of the remote controller 20 is pressed, the main body microphone 81 is maintained in a state capable of collecting sound (active state), and the pressure of the sounds collected from the two microphones 24 and 81 is strong ( Speech recognition of the sound collected from the microphone of the one with the louder sound pressure or the one that picks up the voice clearly (higher intelligibility) (resulting in the higher speech recognition rate) You may make it output to the part 71. FIG.

The clarity of speech can be evaluated by, for example, a clarity index (eg, SII: Speech Intelligibility Index). SII is standardized as "ANSI S3.5-1997", and basically, for each divided frequency band, the signal-to-noise ratio and the coefficient for each frequency (contribution rate to clarity for each frequency) , and sum them up to obtain the overall clarity index.
This may be simplified, and the intelligibility index may be obtained by limiting the frequency band to a frequency band that greatly contributes to speech intelligibility (eg, 1000 Hz to 3,000 Hz).

At this time, it is possible to evaluate whether the speech recognition rate is high or low from either the sound pressure Pv or the articulation index SII.
The speech recognition rate may be evaluated by combining the sound pressure Pv and the articulation index SII. For example, the speech recognition rate can be evaluated by linear addition of the sound pressure Pv and the articulation index SII, as in the following equation (1).
R=K1*Pv+K2*SII... Formula (1)
Here, the coefficients K1 and K2 are proportional coefficients.
That is, the speech with the larger value R determined by the equation (1) can be made to have a higher speech recognition rate.

The IR receiver 18 receives instructions (operation input) from the remote controller 20, such as selection of a channel (broadcasting station), start of recording (REC), playback of a recorded program (PLAY), pause (PAUSE), A command corresponding to special reproduction, menu display, or the like is input to the control module 65 through infrared communication.

The BT communication unit 19 performs Bluetooth communication (short-range wireless communication) with the remote controller 20 . The BT communication unit 19 receives an audio signal transmitted from the remote controller 20 and inputs it to the control module 65 . The BT communication unit 19 functions as a first sound acquisition unit that acquires from the remote controller 20 the first sound collected by the microphone 24 of the remote controller 20 .

In addition, by providing a WiFi (Wireless Fidelity) communication unit or the like, wireless communication may be performed with a short-range wireless communication device conforming to the WiFi standard or the like. Further, a short-range wireless communication unit conforming to a standard such as NFC (Near Field Communication) may be provided to communicate with an external device conforming to the same standard.

The USB I/F 76 communicates data and signals with an external connection device (input device or storage device) conforming to the USB standard. Input devices include, for example, a keyboard and a mouse. The storage device is the HDD 102 connected to the USB terminal as in this example. The HDDs 101 and 102 can use the storage areas in various ways depending on the settings.

The HDD 101 is set to schedule recording or manual recording of programs individually specified by the user from an electronic program guide (EPG). ) and recording by a time shift machine function (all program recording function: also referred to as "all recording function" or "loop recording function") for recording all programs in a predetermined time period for a certain period of time. The reverse setting is also possible.

In this example, the HDD 101 is provided inside the device and the HDD 102 is connected to the outside of the device, but a plurality of externally connected HDDs 102 may be connected.

The communication I/F 73 is controlled by the control module 65 to access the external network NTW and communicate with various service servers on the external network NTW (servers 200, 201, etc. that provide content search services by voice recognition). I do. Specifically, the communication I/F 73 is controlled by the control module 65 to make a search request for obtaining information (transmission of input information), receive search results (obtain information), and the like.

The server 200 manages program information used for watching TV programs, recording reservations, history storage of recorded content, etc., and searches for programs and content related to programs by utterance (voice) of the AI assistant function. Services (hereinafter referred to as "A service", "first search service", etc.) are provided.

The server 201 is a computer that provides search services (hereinafter referred to as "service B", "second search service", etc.) for content on the Internet using speech (voice) as an AI assistant function. It is possible to search a wide range of contents such as Internet programs and dictionaries.

The services of these service servers support not only searches by voice, but also searches by character data obtained by converting voice into characters. Here, the term voice data includes a digital voice signal and its character data.

A control module 65 includes a ROM (read only memory) 66 that stores a control program that controls the operation of the device, a RAM (random access memory) 67 that provides a work area for processing signals and data, recording reservation information, various , a flash memory 68, a setting unit 69, a recording unit 70, a voice recognition unit 71, a control unit 72, etc., which store setting information, control information, etc., and a recording/reproducing apparatus main body including the above-described signal processing, etc. 100 (broadcast reception function, program recording and playback function, setting function, voice function, network communication function) and operations are controlled in an integrated manner. A voice function is a speech recognition function of the speech recognition section 71 including a speech/text conversion function and a parsing function.

As a result, the recording/reproducing apparatus main body 100 receives terrestrial digital broadcasting with the broadcast receiving function, and reproduces programs (video data including audio) recorded on the HDDs 101 and 102 with the recording function, so that the user can enjoy the program. becomes available for viewing. Also, by connecting to a home network, the recording/reproducing apparatus main body 100 can reproduce programs saved (recorded) in another recorder connected to the home network or in a home server.

The flash memory 68 stores a recording reservation table for reserved recording by the reservation recording function, a recording reservation table for individual programs, recording information as attribute information of the recorded programs, setting information of the voice function, and the like. . The setting information may be set in advance, or may be set by a user's selection operation from a setting menu screen displayed by the setting unit 69 . The setting information includes selection conditions for selecting one of search services provided by one or more service servers (servers 200, 201, etc.).

That is, the flash memory 68 stores conditions for making one of the two microphones 24 and 81 active (operating state) or inactive (operating stop state), or two conditions obtained by the two microphones 24 and 81. It can be said that it is a storage unit that stores conditions for using one of the voices.

The setting unit 69 displays a screen for setting setting information in the flash memory 68 , and stores the confirmed setting information in the flash memory 68 after the setting operation by the user.

The recording unit 70 stores the first voice acquired by the BT communication unit 1 (first voice acquisition unit) and the second voice acquired by the input voice processing unit 64 (second voice acquisition unit) in the flash memory 68, the HDD 101, or the like. stored (recorded) in

The voice recognition unit 71 reads the voice recorded by the recording unit 70 from the flash memory 68, the HDD 101, or the like and analyzes it, that is, performs voice recognition processing.

If the recording/reproducing device main body 100 has a high processing capability, the recorded sound is not read out and processed, but the sound (first sound) from the remote controller 20 received by the BT communication unit 26 or the main body microphone 81 is used. The collected sound (second sound) may be analyzed in real time. Analyzing speech means transcribing speech (voice uttered by the user) into text, syntactically analyzing the textualized voice data using a preset analysis dictionary, and extracting words, meaningful characters, or character strings. It refers to speech recognition processing that extracts (keywords).

The control unit 72 inputs the first voice of the microphone 24 of the remote controller 20 and the second voice of the main body microphone 81 that meets a preset condition to the voice recognition unit 71 for voice recognition processing.

Here, the conditions include the following conditions "1." to "3.".
Condition "1."... For example, when a signal is received by operating the voice button 21b of the remote controller 20, the operation of the main body microphone 81 is stopped.
Condition "2.": When a signal is received by operating the voice button 21b of the remote controller 20, the first voice obtained from the remote controller 20 is recognized by the voice recognition unit 71.
Condition 3. Use the better quality of the two recorded voices.
etc.

The control unit 72 calls the control program held in the ROM 66 to the work area provided by the RAM 67, and executes processing corresponding to the input signal and the control signal based on the called control program.

The control unit 72 controls, for example, a recording/playback function and a voice function, and acquires various information (attribute information) related to contents (programs).

The control unit 72 controls each unit (setting unit 69, recording unit 70 , speech recognition unit 71, etc.).

In addition, the control unit 72 writes into the flash memory 68 various setting information and management information related to other recorders and televisions connected to the home server in the home network.

The control unit 72 controls the recording/playback function based on, for example, an operation instruction (control input) from the user or recording reservation information for reserved recording, and outputs video signals, audio signals, etc. Record (record) on (one of the HDDs 101 and 102).

The control unit 72 transmits the characters or character strings as the result of recognition by the speech recognition unit 71 and the acquired speech (first speech or second search for content using voice) and receive the results of the search.

In other words, the control unit 72 sends a search request (transmission of input information) for acquiring content, receives search results (acquisition of content), etc. to the service server (one of the servers 200 and 201). conduct.

More specifically, the control unit 72 provides the service server (one of the servers 200 and 201) via the communication I/F 73 with the character or character string as the result of recognition by the speech recognition unit 71 and the acquired character string. A content search request is made using at least part of the voice, and the search result received from the server in response to the search request is output to the video display unit 14 .

The control unit 72 also transmits and receives information to and from service servers (servers 200, 201, etc.) connected to an external network NTW via the communication I/F 73. FIG. Furthermore, the control unit 72 performs information transmission with USB compatible devices via the USB I/F 76 .

Furthermore, the control unit 72 displays the content (program) of the channel that has been received by the tuner 51 and selected. The control unit 72 also refers to the recording reservation information included in the recording reservation list stored in the flash memory 68 and controls the recording operation of the content (program) based on the signal received by the tuner 51 . The recording operation includes recording by manual operation. The recording destination of the content (program) during the recording operation is, for example, the HDD 101 provided inside the device, the HDD 102 connected via the USB I/F 76, or the like.

2 to 4, the operation corresponding to the above conditions "1." to "3." will be described. First, with reference to the flow chart of FIG. 2, a first operation example of the recording/reproducing apparatus 1 according to the condition "1." will be described.
In the case of this first operation example, when the recording/reproducing apparatus main body 100 is activated, the control unit 72 activates the main body microphone 81 and collects sounds from around the main body microphone 81 (step S101 in FIG. 2).

If no signal is received because the voice button 21b of the remote control 20 is not operated during sound collection by the main body microphone 81 (No in step S102), the control unit 72 controls the recording unit 70 and the voice recognition unit 71. , the voice collected by the main body microphone 81 is recorded (step S103), and the recorded voice is subjected to voice recognition processing (step S104).

Then, the control unit 72 searches for a service server (one of the servers 200 and 201) set in advance as a request destination based on the result of the voice recognition processing (words (characters), character strings, keywords, etc.) and the voice. A request is made (step S105). The search request includes at least a portion of the recorded voice and, if necessary, words and the like resulting from the analysis.

Upon receiving the search request, the service server (server 200 or 201 ) searches for content based on the received voice or word, and sends the search result (content) to the recording/reproducing apparatus main body 100 .

When the recording/reproducing apparatus main body 100 receives the search result (content) transmitted from the server (step S106), the content is output to the video display unit 14 (step S107) and displayed.

On the other hand, when the user operates the button 21 of the remote control 20 while the main body microphone 81 is collecting sound (step S101), the signal processing unit 22 of the remote control 20 generates a signal corresponding to the button 21, and generates a signal corresponding to the button 21. The received signal is transmitted from the IR transmitter 23 .

Here, for example, when the voice button 21b, which is a specific button of the remote controller 20, is pressed, the signal processing unit 22 activates the microphone 24, and the microphone 24 starts collecting sounds.

Here, when the user speaks into the microphone 24 of the remote controller 20 , the voice is collected by the microphone 24 and transmitted from the BT communication section 26 after voice processing.

In the recording/reproducing apparatus main body 100, when the IR signal transmitted from the remote controller 20 is received by the IR receiving section 18 (Yes in step S102), the control section 72 determines whether or not the signal is the signal of the voice button 21b. (step S108).

As a result of determination, if it is not the signal of the voice button 21b (No in step S108), the function corresponding to the signal is controlled (step S109).

On the other hand, if the received signal is the voice button 21b signal (Yes in step S108), then the controller 72 refers to the conditions in the flash memory 68. FIG. The condition "1." for this operation is to stop the operation of the main body microphone 81 when a signal is received by operating the voice button 21b of the remote controller 20. 81 is deactivated (step S110), and the main body microphone 81 stops collecting the second sound.

Then, when the first sound from the remote controller 20 is received (step S111), the control unit 72 controls the recording unit 70 to record the first sound from the remote controller 20 (step S112).

According to this first operation example, the recording/reproducing apparatus main body 100 is provided with the setting unit 69, the recording unit 70, the voice recognition unit 71, and the control unit 72, and when the voice button 21b of the remote control 20 is pressed and the signal is received, By using the first voice acquired from the microphone 24 of the remote controller 20 for voice recognition processing with the main body microphone 81 inactive, the accuracy of voice recognition can be improved.

For example, normally, sound collection and voice recognition processing are performed by the main body microphone 81, and when the voice button 21b of the remote controller 20 is pressed and a trigger signal for starting recording is received, the control unit 72 is triggered by the trigger. By deactivating the main body microphone 81 and activating the microphone 24 of the remote controller 20 and using the first sound collected by the remote controller 20, which is close to the speaker, for speech recognition processing, a story in which the remote controller 20 is operated is performed. It is possible to acquire high-quality speech of a person (user) and perform speech recognition processing with high accuracy.

Next, a second operation example of the recording/reproducing apparatus 1 according to the condition "2." will be described with reference to the flow chart of FIG. In this second operation example, the same reference numerals are given to the same operations as in the first operation example, and the explanation thereof will be omitted.

In the case of this second operation example, when the recording/reproducing apparatus main body 100 is activated, the control section 72 activates the main body microphone 81 and collects sounds from around the main body microphone 81 (step S101 in FIG. 3).

If no signal is received because the voice button 21b of the remote control 20 is not operated during sound collection by the main body microphone 81 (No in step S102), the control unit 72 operates in the same manner as in the first operation example (step S103). ~S107).

On the other hand, when the user operates the button 21 of the remote control 20 while the main body microphone 81 is collecting sound (step S101), the signal processing unit 22 of the remote control 20 generates a signal corresponding to the button 21, and generates a signal corresponding to the button 21. The received signal is transmitted from the IR transmitter 23 .

Here, for example, when the voice button 21b, which is a specific button of the remote controller 20, is pressed, the signal processing unit 22 activates the microphone 24, and the microphone 24 starts collecting sounds.

Then, when the user speaks into the microphone 24 of the remote controller 20 , the voice is collected by the microphone 24 and transmitted from the BT communication section 26 after voice processing.

In the recording/reproducing apparatus main body 100, when the IR signal transmitted from the remote controller 20 is received by the IR receiving section 18 (Yes in step S102), the control section 72 determines whether or not the signal is the signal of the voice button 21b. (step S108).

As a result of determination, if it is not the signal of the voice button 21b (No in step S108), the function corresponding to the signal is controlled (step S109).

On the other hand, if the received signal is the signal of the voice button 21b (Yes in step S108), then the control unit 72 waits for reception of voice from the remote controller 20, and when the voice of the remote controller 20 is received (step S121), and controls the recording unit 70 to record the voice from the remote controller 20 (step S122). Since the main body microphone 81 remains active during this period, the recording of the sound collected by the main body microphone 81 is also continued (step S103).

Next, the controller 72 refers to the conditions of the flash memory 68 . Condition "2." of this operation is a condition that, when a signal is received by operating the voice button 21b of the remote controller 20, the speech recognition unit 71 recognizes the first voice obtained from the remote controller 20. Of the two sounds respectively recorded by the recording unit 70, the control unit 72 inputs the first sound obtained from the remote controller 20 to the voice recognition unit 71, and causes the voice recognition unit 71 to perform voice recognition processing (S123). After that, the operation using the speech recognition result of the speech recognition unit 71 is the same as in the first embodiment.

According to this second operation example, when a signal is received by operating the voice button 21b of the remote controller 20, the control unit 72 outputs two sounds (first sound and second sound) respectively recorded by the recording unit 70. Among them, the recorded first voice of the remote controller 20 is input to the voice recognition section 71, and the voice recognition section 71 performs voice recognition processing.

For example, when the recording start trigger is activation of the recording/reproducing apparatus main body 100 or pressing of the voice button 1b of the remote controller 20, the trigger triggers the recording of the second voice by the main microphone 81 and the microphone 24 of the remote controller 20. simultaneously with the recording of the first voice of If the trigger source is the remote controller 20 that is close to the speaker (user), the voice collected by the microphone 24 of the remote controller 20 is acquired and voice recognition processing is performed. By recognizing a high-quality voice from the remote controller 20 that is close to the speaker from among a plurality of simultaneously recorded voices, the voice recognition accuracy can be improved.

Next, a third operation example of the recording/reproducing apparatus 1 according to the condition "3." will be described with reference to the flow chart of FIG. In this third operation example, the same reference numerals are given to the same operations as in the second operation example, and the explanation thereof will be omitted.

In the case of this third operation example, the operation from the activation of the recording/reproducing apparatus main body 100 to the recording of the sound collected by each microphone is the same as in the second operation example, and the description thereof will be omitted.

The control unit 72 refers to the conditions in the flash memory 68 while each of the two voices is being recorded. Since the condition "3." of this operation is to use the sound with the better quality out of the two recorded sounds, the control unit 72 controls the two sounds recorded by the recording unit 70 respectively. The sound quality is checked, and the sound with the higher speech recognition rate out of the two sound-checked sounds is input to the speech recognition section 71, and the speech recognition section 71 performs speech recognition processing (S131, S132). After that, the operation using the speech recognition result of the speech recognition unit 71 is the same as in the first and second embodiments.

According to this third operation example, the control unit 72 checks the quality of each of a plurality of voices (first voice and second voice) acquired and recorded from the microphone 24 of the remote controller 20 and the main body microphone 81, and checks the quality of each of the plurality of recorded voices. Since the voice with the highest quality among the voices is used for voice recognition processing, the accuracy of voice recognition can be improved.

In this third operation example, the trigger for starting sound collection by the microphone 24 is the same as in the second operation example. It may be when the voice button 21b is pressed, that is, when the signal of the voice button 21b is received.

As described above, according to the recording/reproducing apparatus 1 of this embodiment, both the remote controller 20 (external terminal) and the recording/reproducing apparatus main body 100 (electronic device) are provided with microphones (sound collectors) to collect sound, Out of the collected sounds, by configuring to use the sounds that meet the conditions of "1." A plurality of microphones 24 and 81 can be selectively used depending on the situation, and the sounds collected by the respective microphones 24 and 81 can be utilized.
In addition, in the present embodiment, high-quality voice data can be obtained by selectively using the plurality of microphones 24 and 81 depending on the situation of the speaker, for example, by switching the sound collecting unit to the microphone 24 closer to the speaker. In addition, it is possible to prevent the main body microphone 81 from making an erroneous response while the microphone 24 of the remote controller 20 is collecting sounds.

In the above-described embodiment, an example in which the microphones 24, 81, etc. are provided in the main body 100 of the recording/reproducing device and the remote controller 20 is shown, but a plurality of external terminals (first remote controller and second remote controller) are each provided with a microphone. A plurality of sounds may be transmitted from the remote controller to the recording/reproducing apparatus main body 100 .

That is, the recording/reproducing apparatus main body 100 acquires the first sound collected by the microphone of the first remote controller and the second sound collected by the microphone of the second remote controller, and presets the sound in the recording/reproducing apparatus main body 100. A configuration may be adopted in which voices that meet the conditions are selected and used for voice recognition processing.

While embodiments of the invention have been described, the embodiments have been presented by way of example and are not intended to limit the scope of the invention. This novel embodiment can be embodied in various other forms, and various omissions, replacements, and modifications can be made without departing from the scope of the invention. The above-described embodiments and modifications thereof are included in the scope and gist of the invention, and are included in the scope of the invention described in the claims and equivalents thereof.

Further, each component of the recording/reproducing apparatus 1 shown in the above embodiment may be realized by a program installed in a storage such as a hard disk device of a computer, and the above program may be stored on a computer-readable electronic medium: electronic media. The functions of the present invention may be realized by the computer by storing the program and causing the computer to read the program from the electronic medium.

Examples of electronic media include recording media such as CD-ROMs, flash memories, removable media, and the like. Furthermore, it may be realized by distributing and storing the constituent elements in different computers connected via a network, and by communicating between the computers that operate each constituent element.

DESCRIPTION OF SYMBOLS 1... Recording/reproducing apparatus 14... Video display part 15... Speaker 16... Operation part 18... IR receiving part 19... BT communication part 20... Remote controller (remote controller) 21... Button 21a... Setting button, 21b voice button 22 signal processing unit 23 IR transmission unit 24 microphone 25 audio processing unit 26 BT communication unit 50 antenna 51 tuner 52 OFDM demodulator 53 signal Processing unit 58 Graphic processing unit 59 Audio processing unit 61 OSD signal generation unit 62 Video processing unit 64 Input audio processing unit 65 Control module 68 Flash memory 69 Setting unit 70... Recording unit, 71... Voice recognition unit, 72... Control unit, 73... Communication interface (communication I/F), 76... USB interface (USB I/F), 81... Main unit microphone, 100... Recording/reproducing apparatus main unit, 101, 102... Hard disk drive (HDD), 200, 201... Server, NTW... Network.

Claims (6)

An electronic device wirelessly or wiredly connected to an external terminal having a first sound collecting unit that collects a first sound around the electronic device,
a first sound acquiring unit that acquires from the external terminal a first sound collected by the first sound collecting unit of the external terminal;
a second sound collecting unit that collects a second sound around the self;
a second sound acquisition unit that acquires the second sound collected by the second sound collection unit;
a speech recognition unit that performs speech recognition processing on input speech;
a control unit for inputting, to the speech recognition unit, a speech that matches a preset condition out of the first speech and the second speech, and performing speech recognition processing;
The condition is "based on the sum of the clarity index for each frequency band obtained from the sound pressure and the signal-to-noise ratio for each divided frequency band and the coefficient indicating the contribution rate to the clarity for each frequency band. Use the voice with the higher required voice recognition rate."
The control unit
An electronic device that causes the speech recognition unit to perform speech recognition processing on one of the first speech and the second speech that has a higher speech recognition rate. a recording unit that records the first sound acquired by the first sound acquisition unit and the second sound acquired by the second sound acquisition unit;
The control unit
2. The electronic device according to claim 1, wherein the voice recognition unit recognizes the voice that matches the condition among the first voice and the second voice recorded by the recording unit. A receiving unit that receives an instruction signal transmitted from the external terminal,
3. The electronic device according to claim 2, wherein the recording unit records the first voice obtained from the external terminal when the receiving unit receives a specific instruction signal from the external terminal. . An electronic device wirelessly or wiredly connected to an external terminal having a first sound collecting unit that collects a first sound around the electronic device,
a first sound acquiring unit that acquires from the external terminal a first sound collected by the first sound collecting unit of the external terminal;
a second sound collecting unit that collects a second sound around the self;
a second sound acquisition unit that acquires the second sound collected by the second sound collection unit;
a speech recognition unit that performs speech recognition processing on input speech;
a control unit for inputting, to the speech recognition unit, a speech that matches a preset condition out of the first speech and the second speech, and performing speech recognition processing;
The condition is "Speech obtained based on the sum of the clarity index for each frequency band obtained from the signal-to-noise ratio for each divided frequency band and the coefficient indicating the contribution rate to the clarity for each frequency band. Use the voice with the higher recognition rate"
The control unit
An electronic device that causes the speech recognition unit to perform speech recognition processing on one of the first speech and the second speech that has a higher speech recognition rate. A program for operating an electronic device wirelessly or wiredly connected to an external terminal having a first sound collecting unit that collects a first sound around itself,
the electronic device,
a first sound acquiring unit that acquires from the external terminal a first sound collected by the first sound collecting unit of the external terminal;
a second sound acquisition unit that acquires a second sound collected from around the second sound collection unit by a second sound collection unit provided in the electronic device;
a speech recognition unit that performs speech recognition processing on input speech;
Functioning as a control unit for inputting a sound that matches a preset condition out of the first sound and the second sound to the voice recognition unit and performing voice recognition processing;
The condition is "based on the sum of the clarity index for each frequency band obtained from the sound pressure and the signal-to-noise ratio for each divided frequency band and the coefficient indicating the contribution rate to the clarity for each frequency band. Use the voice with the higher required voice recognition rate."
The control unit
causing the speech recognition unit to perform speech recognition processing on the speech with the higher speech recognition rate out of the first speech and the second speech;
A program characterized by A speech recognition method in an electronic device wirelessly or wiredly connected to an external terminal having a first sound collecting unit that collects a first sound around the self,
acquiring from the external terminal a first sound collected by the first sound collecting unit of the external terminal;
Acquiring a second sound collected from around the second sound collecting unit by a second sound collecting unit provided in the electronic device,
performing voice recognition processing on a voice that matches a preset condition among the first voice and the second voice;
The condition is "based on the sum of the clarity index for each frequency band obtained from the sound pressure and the signal-to-noise ratio for each divided frequency band and the coefficient indicating the contribution rate to the clarity for each frequency band. Use the voice with the higher required voice recognition rate."
The speech recognition processing includes:
voice recognition processing of the voice with the higher voice recognition rate out of the first voice and the second voice;
A speech recognition method characterized by:

Images