JP2023133782A - Speech recognition text display system, speech recognition text display device, speech recognition text display method and program - Google Patents

Abstract

To provide a speech recognition text display system, a speech recognition text display device, a speech recognition text display method, and a program that enable participants to easily understand text information converted from speech signals.SOLUTION: A speech recognition text display system for converting speech signals into text information to display it, comprises: an acquisition unit that acquires speech signals; a speech recognition unit that speech-recognizes speech signals acquired by the acquisition unit and outputs text information; and a display unit that displays the text information. The speech recognition unit calculates recognition likelihood for each homonym for words whose pronunciation is determined by speech recognition to have multiple homonyms in common, and switches whether to output the determined words as text information converted into Chinese characters or text information using kana characters according to the magnitude of each calculated value.SELECTED DRAWING: Figure 1

Description

The present invention relates to a voice recognition text display system, a voice recognition text display device, a voice recognition text display method, and a program.

2. Description of the Related Art Conventionally, there has been known a device that converts the content (audio signals) uttered by each participant into text and displays it when a plurality of participants hold a conference (for example, see Patent Document 1). Such devices are sometimes used to assist hearing-impaired people in attending conferences.

JP2019-179480A

For example, in the device described in Patent Document 1, when a word that has a homophone is included in the content uttered by a conference participant, the word is converted into a kanji that is not intended by the speaker and displayed. There may be cases where In this case, it may become difficult for other participants to understand what the speaker has said. Such problems become particularly noticeable when hearing-impaired people participate in a meeting. This is because hearing-impaired people understand the content of the meeting by reading the text.

The present invention has been made in view of the above problems, and includes a voice recognition text display system, a voice recognition text display device, and a voice recognition text display system that can make it easier for participants to understand text information converted from voice signals. The purpose of this invention is to provide a recognized text display method and program.

(1) In order to achieve the above object, a voice recognition text display system (1) according to one aspect of the present invention is a voice recognition text display system that converts a voice signal into text information and displays the voice signal. an acquisition unit (222) that acquires the audio signal, and a voice recognition unit (speech recognition unit 223, text conversion unit 224, assistant) that performs voice recognition on the audio signal acquired by the acquisition unit and outputs the text information. a reception analysis section 225, a calculation section 226, a conversion switching section 227), and a display section (display section 203, display section 303, minutes creation section 228, image output section 241) that displays the text information; The speech recognition unit calculates a recognition likelihood for each of the plurality of homophones for words for which it is determined in the speech recognition that there are a plurality of homophones having a common pronunciation, and Depending on the magnitude of the recognition likelihood, it is switched whether to output the determined word as the text information converted into Kanji or as the text information in Kana characters.

(2) Furthermore, in the speech recognition text display system according to one aspect of the present invention, the speech recognition unit may detect the determined word when the maximum value of the plurality of calculated recognition likelihoods is lower than a predetermined value. may be output as the text information in kana characters without converting it into kanji.

(3) Furthermore, in the voice recognition text display system according to one aspect of the present invention, when displaying the text information, the display unit displays the determined word and the kana character by the voice recognition unit. The words output as the text information may be displayed in a format different from other words.

(4) In order to achieve the above object, a voice recognition text display device according to one aspect of the present invention is a voice recognition text display device that converts a voice signal into text information and displays it, and acquires the voice signal. The voice recognition unit includes an acquisition unit, a voice recognition unit that performs voice recognition on the voice signal acquired by the acquisition unit and outputs the text information, and a display unit that displays the text information. The unit calculates the recognition likelihood for each of the plurality of homonyms with respect to the word for which it is determined in the speech recognition that there are a plurality of homonyms with a common pronunciation, and Depending on the magnitude of the likelihood, it is switched whether the determined word is output as the text information converted into kanji or as the text information in kana characters.

(5) In order to achieve the above object, a voice recognition text display method according to one aspect of the present invention is a voice recognition text display method in a voice recognition text display system that converts a voice signal into text information and displays it, comprising: an acquisition step in which an acquisition unit acquires a voice signal; a voice recognition step in which a voice recognition unit performs voice recognition on the voice signal acquired by the acquisition unit and outputs the text information; and a display unit. and a displaying step of displaying the text information, and in the speech recognition step, the speech recognition unit is configured to display information about a word for which it is determined in the speech recognition that there are a plurality of homophones having a common pronunciation. , calculate the recognition likelihood for each of the plurality of homophone opposition words, and output the determined word as the text information converted into kanji according to the magnitude of the plurality of calculated recognition likelihoods. or outputting the text information in kana characters.

(6) In order to achieve the above object, a program according to one aspect of the present invention includes an acquisition step of acquiring an audio signal, and an acquisition section for acquiring an audio signal in a speech recognition text display system that converts an audio signal into text information and displays the result. A voice recognition step of performing voice recognition on the voice signal acquired in the step and outputting the text information, and a display step of displaying the text information, and in the voice recognition step, the voice For a word that is determined to have a plurality of homonyms with a common pronunciation in recognition, the recognition likelihood for each of the plurality of homonyms is calculated, and the magnitude of the plurality of calculated recognition likelihoods is calculated. The determined word is output as the text information converted into kanji or as the text information in kana characters.

According to the above-mentioned (1), (4), (5), or (6), there is an effect that the participants can easily understand the text information converted from the audio signal.

According to (2) above, it is possible to reduce the possibility that text information converted into kanji characters not intended by the speaker will be displayed.
According to (3) above, the participant can identify words that were displayed as text information in kana characters because the recognition likelihood of homophones was low.

1 is a block diagram showing a configuration example of a speech recognition text display device (speech recognition text display system) according to the present embodiment. FIG. FIG. 3 is a diagram showing an example of an image displayed on the display unit of the parent device according to the present embodiment. FIG. 3 is a diagram illustrating an example of an image displayed on a display unit of a slave device according to the present embodiment. 2 is a flowchart illustrating an example of a processing procedure performed by the speech recognition text display device (speech recognition text display system) according to the present embodiment.

Embodiments of the present invention will be described below with reference to the drawings.

First, an example of a situation in which the voice recognition text display device (voice recognition text display system) of this embodiment is used will be described.
The voice recognition text display device (voice recognition text display system) of this embodiment is used in a conference attended by two or more people. Among the participants, a person with a speech disability may be participating in the conference. Participants who can speak should each wear a microphone or use a terminal (smartphone, tablet terminal, personal computer, etc.) equipped with a microphone. Hearing-impaired participants will use a terminal that can input text. The voice recognition text display device recognizes voice signals uttered by participants, converts them into text, and displays the text on their respective terminals. Furthermore, the voice recognition text display device displays text information input by hearing-impaired people on their own terminals.

FIG. 1 is a block diagram showing a configuration example of a speech recognition text display device (speech recognition text display system) 1 according to the present embodiment.
As shown in FIG. 1, a voice recognition text display device (voice recognition text display system) 1 includes a base unit 2, a slave unit 3a, a slave unit 3b, . . . . Note that if one of the handset 3a, handset 3b, . . . is not specified, it will simply be referred to as the handset 3.
The base unit 2 and the slave unit 3 are connected via a wired or wireless network 4.

The base device 2 includes a sound collection section 201, an operation section 202, a display section 203, a communication section 204, an authentication section 211, an acoustic model/dictionary storage section 221, an acquisition section 222, a speech recognition section 223, a text conversion section 224, and a modification section. It includes an analysis section 225, a minutes creation section 228, a minutes storage section 229, a text acquisition section 231, and an image output section 241. The dependency analysis section 225 includes a calculation section 226 and a conversion switching section 227. However, the calculation unit 226 or the conversion switching unit 227 may be provided separately from the dependency analysis unit 225.

The handset 3 includes a sound collection section 301, an operation section 302, a display section 303, a communication section 304, and a processing section 305. The sound collection section 301, the operation section 302, the display section 303, the communication section 304, and the processing section 305 are connected via a bus 306.

<Slave unit 3>
First, the handset 3 will be explained.
The handset 3 is, for example, a smartphone, a tablet terminal, a personal computer, or the like. Note that the handset 3 may include an audio output unit, a motion sensor, a GPS (Global Positioning System), and the like.

The sound collection unit 301 is a microphone. The sound collection unit 301 collects a user's audio signal, converts the collected audio signal from an analog signal to a digital signal, and outputs the audio signal converted to the digital signal to the processing unit 305.

The operation unit 302 detects a user's operation and outputs the detected result to the processing unit 305. The operation unit 302 is, for example, a touch panel sensor provided on the display unit 303, a keyboard with priority connection or wireless connection, or the like.

The processing unit 305 generates setting information based on the operation result of the operation unit 302 and outputs the generated setting information to the communication unit 304. Here, the setting information includes participant identification information. The setting information may include information indicating whether the sound collection section is used or not, and information indicating whether or not the operation section is used. The processing unit 305 generates a login instruction based on the operation result of the operation unit 302 and outputs the generated login instruction to the communication unit 304. Here, the login instruction includes identification information of the participant and identification information of the handset 3. The processing unit 305 adds identification information to text information based on the operation result of the operation unit 302 and outputs the text information to the communication unit 304 . The processing unit 305 adds identification information to the audio signal output by the sound collection unit 301 and outputs the audio signal to the communication unit 304 . The processing unit 305 acquires the image data output by the communication unit 304 and outputs the acquired image data to the display unit 303. The processing unit 305 establishes communication with the base device 2 based on the information output by the communication unit 304 that permits login. When the processing unit 305 receives a speech restriction instruction (input restriction instruction) from the base device 2, the processing unit 305 may restrict text input. Further, when the processing unit 305 receives a speech restriction instruction from the base device 2, the processing unit 305 may also restrict voice input.

The display unit 303 displays the image data output by the processing unit 305. The display unit 303 is, for example, a liquid crystal display device, an organic EL (electroluminescence) display device, an electronic ink display device, or the like. Note that the image displayed on the display unit 303 will be described later.

The communication unit 304 transmits the setting information output by the processing unit 305 to the base device 2 via the network 4. The communication unit 304 transmits the login instruction output by the processing unit 305 to the base device 2 via the network 4. The communication unit 304 transmits text information or audio signals output by the processing unit 305 to the base unit 2 via the network 4. Note that the text information or audio signal to be transmitted includes the identification information of the user and the identification information of the handset 3. The communication unit 304 receives the image data transmitted by the base device 2 and outputs the received image data to the processing unit 305. When the communication unit 304 receives the information that permits login transmitted from the base device 2, the communication unit 304 outputs the received information that permits login to the processing unit 305.

<Main unit 2>
Next, the base unit 2 will be explained.
The parent device 2 is, for example, a notebook computer.

The sound collection unit 201 is a microphone. The sound collection unit 201 collects a user's audio signal, converts the collected audio signal from an analog signal to a digital signal, and outputs the audio signal converted to the digital signal to the acquisition unit 222.

The operation unit 202 detects a user's operation and outputs the detected result to the text acquisition unit 231. The operation unit 202 is, for example, a touch panel sensor provided on the display unit 203 or a keyboard. The operation unit 202 outputs the result of detecting the operation to the authentication unit 211 during the login process.

The display unit 203 is, for example, a liquid crystal display device, an organic EL display device, an electronic ink display device, or the like. The display unit 203 displays image data output by the image output unit 241. Note that the image displayed on the display unit 203 will be described later.

The communication unit 204 receives the audio signal transmitted by the handset 3 and outputs the received audio signal to the acquisition unit 222. The communication unit 204 receives the text information transmitted by the handset 3 and outputs the received text information to the text acquisition unit 231. The communication unit 204 receives the login instruction sent by the handset 3 and outputs the received login instruction to the authentication unit 211. The communication unit 204 transmits the image data output by the image output unit 241 to the slave unit 3 via the network 4. The communication unit 204 transmits the information output by the authentication unit 211 to permit login to the slave device 3 via the network 4 .

The authentication unit 211 determines whether to permit login based on the identification information of the participant and the identification information of the handset 3 included in the login instruction output by the communication unit 204. When permitting login, the authentication section 211 outputs information permitting login to the communication section 204. The authentication unit 211 determines whether or not to permit the user of the base device 2 to log in, based on the result of the operation on the operation unit 202. When permitting login, the authentication section 211 outputs information permitting login to each functional section, and permits the operation of each functional section. Note that each functional unit includes a communication unit 204, an authentication unit 211, an acoustic model/dictionary storage unit 221, an acquisition unit 222, a speech recognition unit 223, a text conversion unit 224, a dependency analysis unit 225, a calculation unit 226, and a conversion switch. section 227 , a minutes creation section 228 , a minutes storage section 229 , a text acquisition section 231 , and an image output section 241 .

The acoustic model/dictionary storage unit 221 stores, for example, acoustic models, language models, word dictionaries, and the like. The acoustic model is a model based on sound feature amounts, and the language model is a model of information about words and how they are arranged. Further, the word dictionary is a dictionary with a large number of vocabulary words, for example, a large vocabulary word dictionary. Note that the base unit 2 may store and update words and the like that are not stored in the acoustic model/dictionary storage unit 221. Note that the acoustic model/dictionary storage unit 221 may include a DB (database) for each conference, for example. For example, the first DB may be used for general conferences, the second DB may be used for presentations, and the third DB may be used for international conferences. By using a DB tailored to the conference in this way, it becomes easier to appropriately convert homonyms and the like.

The acquisition unit 222 acquires the audio signal output by the sound collection unit 201 or the audio signal output by the communication unit 204, and outputs the acquired audio signal to the voice recognition unit 223.

The voice recognition unit 223 acquires the voice signal output by the acquisition unit 222. The speech recognition unit 223 detects the speech signal of the speech section from the speech signal. The speech section is detected by detecting, for example, an audio signal equal to or higher than a predetermined threshold value as the speech section. Note that the speech recognition unit 223 may detect the utterance section using other well-known techniques. The speech recognition unit 223 refers to the acoustic model/dictionary storage unit 221 and performs speech recognition on the speech signal of the detected speech section using a well-known method. Note that the speech recognition unit 223 performs speech recognition using, for example, the method disclosed in Japanese Patent Application Publication No. 2015-64554. The speech recognition section 223 outputs the recognized recognition result and speech signal to the text conversion section 224. Note that the speech recognition unit 223 outputs the recognition result and the speech signal in association with each other, for example, for each sentence, for each utterance section, or for each utterance.
Note that when audio signals are input simultaneously, the audio recognition unit 223 performs audio recognition for each sound collection unit (201 or 301), for example, by time-sharing processing. Furthermore, when the microphone is a microphone array, the speech recognition unit 223 also performs well-known speech recognition processing such as sound source separation processing, sound source localization processing, and sound source identification processing.

The text conversion unit 224 converts the recognition result output by the speech recognition unit 223 into text by referring to the acoustic model/dictionary storage unit 221. Note that the text information includes information of at least one character. The text conversion unit 224 outputs the converted text information and the acquired audio signal to the dependency analysis unit 225. Note that the text conversion unit 224 may delete interjections such as "ah", "um", "um", and "well" from the result of recognizing the utterance information and convert it into text.

The dependency analysis unit 225 performs morphological analysis and dependency analysis on the text information output by the text conversion unit 224 or the text information output by the communication unit 204 with reference to the acoustic model/dictionary storage unit 221. Note that SVM (Support Vector Machines) is used in the dependency analysis, for example, in the shift-reduce method, the spanning tree method, and the stepwise application method of chunk identification.
The dependency analysis unit 225 outputs text information subjected to morphological analysis and dependency analysis, and the results of the analysis to the minutes creation unit 228. Note that when the dependency analysis unit 225 acquires the audio signal output by the text conversion unit 224, it also outputs the acquired audio signal to the minutes creation unit 228.
Here, in addition to (or instead of) the above-mentioned information, the dependency analysis unit 225 outputs text information processed by the calculation unit 226 and conversion switching unit 227 described below to the minutes creation unit 228. .

The calculation unit 226 calculates the recognition likelihood for each of the plurality of homonyms for words that are determined to have a plurality of homonyms with a common pronunciation in speech recognition (morphological analysis and dependency analysis). . For example, the word "kosei" has multiple homophones that have the same pronunciation, such as "structure," "made of steel," "welfare," "fairness," and so on. The calculation unit 226 refers to the acoustic model/dictionary storage unit 221 or the context before and after “Kousei” in the text information and other information to calculate each of “composition”, “made of steel”, “welfare”, “fairness”, etc. Calculate the recognition likelihood for The calculation unit 226 outputs the plurality of recognition likelihoods obtained by the calculation to the conversion switching unit 227.

The conversion switching unit 227 converts the word that has been determined to have a plurality of homophones with the same pronunciation into the dependency analysis unit 225 according to the magnitude of the plurality of recognition likelihoods output by the calculation unit 226. switches between outputting text information converted to kanji or text information using kana characters. Note that "kana characters" include hiragana and katakana.
For example, the conversion switching unit 227 converts the word ``Kousei'' into ” or “Kosei”, or “Composition” or “Steel” etc.
As an example, when the maximum value of the plurality of recognition likelihoods outputted by the calculation section 226 is lower than a predetermined value, the conversion switching section 227 selects a word that is determined to have a plurality of homophones with a common pronunciation. (“Kosei”) may be output as text information in kana characters without having the dependency analysis unit 225 convert it into kanji.

The minutes creation unit 228 creates minutes for each speaker based on the text information output by the dependency analysis unit 225 or the text acquisition unit 231. The minutes creation unit 228 causes the minutes storage unit 229 to store the audio signal corresponding to the created minutes. Further, the minutes creation unit 228 outputs the created minutes to the image output unit 241. Note that the minutes creation unit 228 may create the minutes by deleting interjections such as "ah", "um", "um", and "well".

The minutes storage unit 229 stores minutes and audio signals in association with each other.

The text acquisition unit 231 acquires the operation result output by the operation unit 202 or the operation result of the operation unit 302 output by the communication unit 204, and generates text information based on the acquired result. The text acquisition unit 231 outputs the generated text information to the minutes creation unit 228.

The image output unit 241 acquires the minutes information output by the minutes creation unit 228. The image output unit 241 generates image data based on the minutes information and outputs the generated image data to the display unit 203 and the communication unit 204.

<Display image of base unit 2>
Next, an example of an image displayed on the display unit 203 of the base device 2 will be described.
FIG. 2 is a diagram showing an example of an image displayed on the display unit 203 of the base device 2 according to the present embodiment.
Image g10 is an image displayed on display unit 203 of base device 2.

Area g100 is an area where participant information is edited.
Area g101 is an area for participant information. The code g102 is the name of the participant. Reference numeral g103 is an icon indicating that the participant inputs text using the operation unit 202 of the base unit 2 or the operation unit 302 of the slave unit 3. Reference numeral g104 is an icon indicating that the participant speaks using the sound collection unit 201 of the base unit 2 or the sound collection unit 301 of the slave unit 3. The code g105 is the number (or identification information) of the microphone used by the participant.

Area g200 is an area for displaying minutes. Note that FIG. 2 shows the state after login. Symbol g201 is a login/logout button image. Reference numeral g202 is a start/end button image of the voice recognition text display device (voice recognition text display system) 1. Reference numeral g203 is a display that lights up while the voice recognition text display device (voice recognition text display system) 1 is in use. Reference numeral g204 is a button image for displaying the minutes stored in the minutes storage section 229 and reproducing the audio signal. Reference numeral g205 is a button image through which the user of the base unit 2 selects whether or not to use the sound collection section 201.

The code g211 is text information input by the first participant by operating the operation unit (202 or 302). The symbol g212 is a pictogram input by the first participant by operating the operation unit (202 or 302). The code g213 is information indicating the date and time when the first participant inputted the text information and pictograms. Code g214 is the name of the first participant.

The code g221 is text information input by the second participant by operating the operation unit (202 or 302). The code g222 is text information obtained by voice recognition of the content uttered by the second participant. The symbol g223 is an icon indicating that text has been input by operating the operation unit (202 or 302). The symbol g224 is an icon indicating that speech has been input by the sound collection unit (201 or 301). The code g231 is text information obtained by voice recognition of the content uttered by the third participant.

Symbol g241 is an example of displaying text information in kana characters for the word "Kosei" uttered by the second participant, according to the recognition likelihood of each homophone "structure", "made of steel", etc.
As shown in the illustrated example, the display unit 203 indicates a word that has been determined to have multiple homophones with the same pronunciation in speech recognition (“Kosei”), and which has been determined by the dependency analysis unit 225 to have multiple homophones with the same pronunciation. Words output as text information in kana characters may be displayed in a format different from other words. Examples of ways to vary the format include italics, boldface, underlining, marker display, font color, font size, font, and other aspects, as well as combinations thereof. In the illustrated example, "Kosei" is displayed in a different format from other text information by making "Kosei" bold and italicized.

Note that in order for the display unit 203 to display the above, the image output unit 241 generates image data in which the word is displayed in a format different from that of other words.

Note that the image shown in FIG. 2 is an example, and the image displayed on the display unit 203 is not limited to this. For example, the display unit 203 displays a word that has been determined to have multiple homophones with the same pronunciation in speech recognition (“Kousei”), and is displayed as text information in kana characters by the dependency analysis unit 225. The output word does not have to be displayed in a different format from other words.

<Display screen of handset 3>
Next, an example of an image displayed on the display unit 303 of the handset 3 will be described.
FIG. 3 is a diagram showing an example of an image displayed on the display unit 303 of the handset 3 according to the present embodiment.
Image g30 is an image displayed on display unit 303 of handset 3.

Area g300 is an area for displaying minutes. The code g311 is text information input by the first participant by operating the operation unit (202 or 302). The code g321 is text information input by the second participant by operating the operation unit (202 or 302). The code g322 is text information obtained by voice recognition of the content uttered by the second participant. The code g331 is text information obtained by voice recognition of the content uttered by the third participant. Area g301 is an area for a text input section. Note that the operation unit 302 may be a software keyboard displayed on the display unit 303, or may be connected to the handset 3 by wire or wirelessly.

Similar to the symbol g241 in FIG. 2, the symbol g341 is an example of displaying the word "Kousei" uttered by the second participant as text information in kana characters. As shown in the illustrated example, the display section 303, like the display section 203, is a word ("kosei") that has been determined to have multiple homophones with a common pronunciation in speech recognition, and , the word output as text information in kana characters by the dependency analysis unit 225 may be displayed in a format different from that of other words. Note that in order for the display unit 303 to display the above, the image output unit 241 generates image data in which the word is displayed in a format different from that of other words.

Note that the image shown in FIG. 3 is an example, and the image displayed on the display unit 303 is not limited to this. For example, the display unit 303 displays a word that has been determined to have multiple homophones with the same pronunciation in speech recognition (“Kousei”), and is displayed as text information in kana characters by the dependency analysis unit 225. The output word does not have to be displayed in a different format from other words.

<Processing performed by voice recognition text display device (voice recognition text display system) 1>
Next, an example of a processing procedure performed by the voice recognition text display device (voice recognition text display system) 1 will be described. FIG. 4 is a flowchart illustrating an example of a processing procedure performed by the voice recognition text display device (voice recognition text display system) 1 according to the present embodiment.

(Step S1) The authentication unit 211 performs a login process based on the operation content of the operation unit (202 or 302). For example, when each user operates the operation unit (202 or 302) and inputs identification information (user ID) that identifies the user and a password, the authentication unit 211 inputs the input identification information and password. Login processing is performed based on

(Step S2) When the user performs input using the sound collection unit (201 or 301), the acquisition unit 222 acquires the audio signal output by the sound collection unit 201 or the communication unit 204, and performs voice recognition on the acquired audio signal. 223.

(Step S3) The voice recognition unit 223 acquires the voice signal output by the acquisition unit 222, and performs voice recognition processing on the acquired voice signal.

(Step S4) The text conversion unit 224 performs text conversion processing on the voice recognition result.

(Step S5) The dependency analysis unit 225 performs dependency analysis and morphological analysis processing for each speaker on the converted text information.

(Step S6) The dependency analysis unit 225 refers to the acoustic model/dictionary storage unit 221 to determine whether or not there are multiple homophones with the same pronunciation for each word included in the analyzed text information. Determine whether If the dependency analysis unit 225 determines that there are multiple homophones (step S6; YES), the process of step S7 is performed, and if it determines that there are no multiple homonyms. (Step S6; NO), the process of step S11 is performed.

(Step S7) The calculation unit 226 calculates the recognition likelihood for each homophone for the word that is determined to have a plurality of homophones with the same pronunciation, and outputs the calculation result to the conversion switching unit 227. do.

(Step S8) The conversion switching unit 227 determines whether the maximum value of the calculated values of the plurality of recognition likelihoods is less than or equal to a predetermined value. If the conversion switching unit 227 determines that the maximum value is less than or equal to the predetermined value (step S8; YES), the process of step S9 is performed, and if it determines that the maximum value exceeds the predetermined value (step S8; If NO), the process of step S10 is performed.

(Step S9) The dependency analysis unit 225 outputs words for which it has been determined that there are multiple homophones with the same pronunciation to the minutes creation unit 228 as text information in kana characters.

(Step S10) The dependency analysis unit 225 converts the word that has been determined to have multiple homophones with the same pronunciation into the homophone (kanji) with the highest recognition likelihood as text information. It is output to the recording section 228.

(Step S11) The dependency analysis unit 225 outputs words for which it has been determined that there are no homophones with the same pronunciation to the minutes creation unit 228 as text information in Kanji. Note that words that cannot be converted into kanji are output to the minutes creation unit 228 as text information in kana characters.

(Step S12) When the user performs an input using the operation unit (202 or 302), the text acquisition unit 231 acquires the operation result output by the operation unit 202 or the communication unit 204, and extracts text information based on the acquired result. It is generated and output to the minutes creation section 228.

(Step S13) The minutes creation unit 228 creates minutes based on the text information output by the dependency analysis unit 225 or the text acquisition unit 231, and outputs the minutes to the image output unit 241.

(Step S14) The image output unit 241 generates an image to be displayed on the display unit (203 or 303) based on the minutes output by the minutes creation unit 228, and outputs it to the display unit 203 or communication unit 204. .

(Step S15) The display unit (203 or 303) displays the image output by the image output unit 241.

The voice recognition text display device (voice recognition text display system) 1 repeats the processes of steps S2 to S15.
Note that the process in FIG. 4 is an example, and the process is not limited to this.

As described above, in this embodiment, a word that has a plurality of homophones with a common pronunciation is intentionally displayed in kana characters according to the recognition likelihood of each homophone.
As a result, according to the present embodiment, it is possible to reduce the possibility that text information converted into kanji that is not intended by the speaker will be displayed. Thereby, according to this embodiment, it is possible to make it easier for participants to understand text information converted from an audio signal.

In the above example, the voice recognition text display device (voice recognition text display system) 1 can input text using the operation unit (202 or 302) and input text using voice recognition using the sound collection unit (201 or 301). However, this is not limited to. For example, the voice recognition text display device (voice recognition text display system) 1 may only allow text input using voice recognition by the sound collection unit (201 or 301).

Further, in the example described above, the voice recognition text display device 1 includes the base unit 2 and a plurality of slave units 3, but the present invention is not limited to this. For example, the voice recognition text display device 1 may include only one slave device 3, or the voice recognition text display device 1 may not include any slave device 3.

Additionally, the authentication section 211, acoustic model/dictionary storage section 221, acquisition section 222, speech recognition section 223, text conversion section 224, dependency analysis section 225, calculation section 226, conversion switching section 227, minutes creation section 228, minutes The record storage section 229, the text acquisition section 231, and the image output section 241 may each be included in the slave device 3. Similarly, the processing unit 305 may be included in the base device 2.

Further, each functional unit of the voice recognition text display device 1 may be provided in a device other than the base device 2 and the slave device 3. Alternatively, each functional unit of the voice recognition text display system 1 may not be provided in the base unit 2 or slave unit 3 or other physical device, but may be provided on one or more servers or the cloud. . Note that each functional unit includes a communication unit 204, an authentication unit 211, an acoustic model/dictionary storage unit 221, an acquisition unit 222, a speech recognition unit 223, a text conversion unit 224, a dependency analysis unit 225, a calculation unit 226, and a conversion switch. 227 , a minutes creation section 228 , a minutes storage section 229 , a text acquisition section 231 , an image output section 241 , a communication section 304 , and a processing section 305 .

Note that a program for realizing all or part of the functions of the speech recognition text display device (speech recognition text display system) 1 of the present invention is recorded on a computer-readable recording medium, and the program is recorded on this recording medium. All or part of the processing performed by the speech recognition text display device (speech recognition text display system) 1 may be performed by loading a program into a computer system and executing it. Note that the "computer system" herein includes hardware such as an OS and peripheral devices. Furthermore, the term "computer system" includes a WWW system equipped with a home page providing environment (or display environment). Furthermore, the term "computer-readable recording medium" refers to portable media such as flexible disks, magneto-optical disks, ROMs, and CD-ROMs, and storage devices such as hard disks built into computer systems. Furthermore, "computer-readable recording medium" refers to volatile memory (RAM) inside a computer system that serves as a server or client when a program is transmitted via a network such as the Internet or a communication line such as a telephone line. This also includes programs that are retained for a certain period of time.

Further, the program may be transmitted from a computer system storing the program in a storage device or the like to another computer system via a transmission medium or by a transmission wave in a transmission medium. Here, the "transmission medium" that transmits the program refers to a medium that has a function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line. Moreover, the above-mentioned program may be for realizing a part of the above-mentioned functions. Furthermore, it may be a so-called difference file (difference program) that can realize the above-described functions in combination with a program already recorded in the computer system.

Although the mode for implementing the present invention has been described above using embodiments, the present invention is not limited to these embodiments in any way, and various modifications and substitutions can be made without departing from the gist of the present invention. can be added.

1... Speech recognition text display device (speech recognition text display system) 203... Display section 222... Acquisition section 224... Text conversion section 225... Dependency analysis section 226... Calculation section 227... Conversion switching section 241... Image output section 303... Display Department

Claims (6)

A voice recognition text display system that converts voice signals into text information and displays it,
an acquisition unit that acquires the audio signal;
a voice recognition unit that performs voice recognition on the voice signal acquired by the acquisition unit and outputs the text information;
a display unit that displays the text information,
The speech recognition unit calculates a recognition likelihood for each of the plurality of homophones with respect to a word for which it is determined in the speech recognition that there are a plurality of homophones with a common pronunciation, and switching whether to output the determined word as the text information converted into kanji or as the text information in kana characters, depending on the magnitude of the plurality of recognition likelihoods;
Speech recognition text display system. The speech recognition unit outputs the text information in kana characters without converting the determined word into kanji when the maximum value of the plurality of calculated recognition likelihoods is lower than a predetermined value.
The speech recognition text display system according to claim 1. When displaying the text information, the display unit displays the determined word and the word output in the text information in kana characters by the speech recognition unit in a format different from other words. indicate,
The voice recognition text display system according to claim 1 or 2. A voice recognition text display device that converts voice signals into text information and displays it,
an acquisition unit that acquires the audio signal;
a voice recognition unit that performs voice recognition on the voice signal acquired by the acquisition unit and outputs the text information;
a display unit that displays the text information,
The speech recognition unit calculates a recognition likelihood for each of the plurality of homophones with respect to a word for which it is determined in the speech recognition that there are a plurality of homophones with a common pronunciation, and switching whether to output the determined word as the text information converted into kanji or as the text information in kana characters, depending on the magnitude of the plurality of recognition likelihoods;
Voice recognition text display device. A voice recognition text display method in a voice recognition text display system that converts a voice signal into text information and displays it,
an acquisition step in which the acquisition unit acquires the audio signal;
a voice recognition step in which a voice recognition unit performs voice recognition on the voice signal acquired by the acquisition unit and outputs the text information;
a display step for displaying the text information;
In the speech recognition step, the speech recognition unit calculates a recognition likelihood for each of the plurality of homophone dissimilar words for the word for which it is determined in the speech recognition that there are a plurality of homophones with a common pronunciation. and switching, depending on the magnitude of the plurality of calculated recognition likelihoods, whether to output the determined word as the text information converted into kanji or as the text information in kana characters.
Speech recognition text display method. A voice recognition text display system that converts voice signals into text information and displays it.
an acquisition step of acquiring an audio signal;
a voice recognition step of performing voice recognition on the voice signal acquired in the acquisition step and outputting the text information;
performing a display step of displaying the text information;
In the speech recognition step, for words for which it is determined in the speech recognition that there are a plurality of homophones with a common pronunciation, a recognition likelihood is calculated for each of the plurality of homophones, and the recognition likelihood is calculated for each of the plurality of homophones. switching whether to output the determined word as the text information converted into kanji or as the text information in kana characters, depending on the magnitude of the plurality of recognition likelihoods;
program.

Images