JP2000076041A - Voice recognizing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a voice recognizing device capable of automatically reflecting the display style of voice recognized sentences with the taste of recording person or operator without performing any complicated setting. SOLUTION: This voice recognizing device is provided with a digital audio data fetching part 1 for fetching an audio file containing audio data and ID information, operating part 2 for setting a parameter such as font or page setting related to the display style of sentences and relating it with the ID information, voice recognition executing and control part 3 for recognizing the voice of audio data in the audio file fetched by the digital audio data fetching part 1, converting it into a sentence, detecting the ID information of the audio file and controlling the display style of voice recognized sentences based on the parameter related to that ID information, and display part 4 for displaying the voice recognized sentences according to the display style controlled by this voice recognition executing and control part 3.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a speech recognition device, and more particularly, to a speech recognition device for recognizing speech data and converting it into a sentence.

[0002]

2. Description of the Related Art Conventionally, dictation contents are written in a dictation by a recorder using a tape recorder or the like, and a transcriptist (also called a transclaver) is recorded on the dictated tape. ), A transcriptionist would play back the tape and document the audio heard using a typewriter or word processor.

[0003] For example, at a work site such as an office, a plurality of recorders generally request a transcript of dictation contents from a single transcriptionist. Due to the large number of people, there were cases where multiple transcriptionists themselves were prepared.

[0004] On the other hand, speech recognition technology has been developed today. For example, by taking dictated speech data into a personal computer and executing speech recognition software, a task of writing the dictated content into sentences has been performed. I have.

[0005] Such speech recognition software includes:
2. Description of the Related Art A device that recognizes voice data input as an analog signal to a personal computer by a microphone or the like has been commercialized.

Further, there have been proposed techniques devised so that voice input can be easily performed in various environments.
For example, there is “ViaVoice Transcription” which is jointly developed and commercialized by IBM and sold by the present applicant.

[0007] This voice recognition software can easily record a voice, for example, on the go, using a portable digital voice recording device as proposed and commercialized by the present applicant. The present invention focuses on capturing a digital voice file later into a personal computer, recognizing the digital voice file using the voice recognition software, and converting the file into a sentence.

Since the sentence recognized by the speech recognition software is converted into text data, the subsequent full-scale processing such as layout and format setting is performed by word processing software or the like. . Therefore, consideration has been given so that the text data subjected to the voice recognition can be easily transferred from the voice recognition software to the word processing software.

[0009] Now that it is becoming common for one person to occupy and use one personal computer, when performing a transcript operation using a speech recognition device, the recorder must be able to use a speech recognition device. It is not uncommon to become an operator of a device (ie, a personal computer with voice recognition software, etc.).

However, since the current speech recognition technology is not perfect, errors may occur in the recognition result, and it is necessary to correct such errors. For this purpose, a speech recognition operator who specializes in correcting errors after performing speech recognition and finishing up a well-formed document should undertake the work of writing dictation recordings by multiple recorders. Can be considered.

In such a case, a case where a font of a sentence to be printed is specified at the request of the recorder or a case where predetermined contents such as a name and an address are added when the document is a letter or a facsimile. And so on.

Since it is considered that it is unlikely that such a specified content is recorded in voice along with the dictation content and transmitted to the voice recognition operator, even if an instruction regarding the arrangement is recorded, it is automatically recognized. There is no established technology that reflects the appearance of text.

In the end, the speech recognition operator transfers the sentence obtained by the speech recognition to a word processor (or word processing software operating on a personal computer), and sets the page settings and fonts registered in advance in the word processor. Work to select the one that matches the preference of the recorder from among, formats, templates, page headers and footers, fixed phrases, logos, illustrations, etc., or to enter it again from the keyboard, to prepare the appearance as a document Is performed, and the text in the display mode obtained as a result is printed or the like.

In the conventional word processing software operating on a personal computer, there has been conventionally proposed one that automatically sets the environment setting of the personal computer in association with a logged-on user.

[0015]

In a conventional speech recognition apparatus using the above-described conventional speech recognition software or word processing software, a display mode suitable for the needs and preferences of a recorder or an operator (speech recognition operator). In order to reflect this in the text that has been subjected to speech recognition, it is necessary to manually perform each setting for each operation of speech recognition, and it cannot be said that the operability is good.

In particular, when there are a plurality of recorders or when there are a plurality of voice recognition operators, there is a strong need for a technique for improving such operability, and its practical application is awaited.

The present invention has been made in view of the above circumstances, and it is possible to automatically reflect the preference of a recorder or an operator in a display mode of a sentence recognized by voice, thereby improving work efficiency. It is an object of the present invention to provide a voice recognition device capable of performing the above.

[0018]

In order to achieve the above object, a voice recognition apparatus according to a first aspect of the present invention performs voice recognition on voice data and additional information related to the voice data added to the voice data, thereby forming a sentence. A speech recognition device for converting a sentence into text data includes reading means for reading the additional information, and control means for controlling a display mode of the converted text based on the additional information.

Further, the speech recognition apparatus according to the second aspect of the present invention
In the voice recognition device according to the first aspect, when the control unit does not have the additional information or is invalid, the control unit controls a display mode of the converted sentence based on information on a predetermined display mode. Is what you do.

Further, the speech recognition apparatus according to a third aspect of the present invention is the speech recognition apparatus according to the first aspect, wherein the control means includes setting means for setting a parameter relating to a display mode of a sentence based on the additional information. , And controls the display mode of the converted text based on the parameters.

The speech recognition apparatus according to a fourth aspect of the present invention is the speech recognition apparatus according to the third aspect, wherein the setting means sets one or more parameters based on the additional information. Of the parameters to be
Parameters not associated with the additional information are set to predetermined values.

In addition, the speech recognition apparatus according to a fifth aspect of the present invention is the speech recognition apparatus according to the first aspect of the present invention, wherein the control means controls the conversion of the sentence converted based on the user ID information in the additional information. This controls the display mode.

[0023]

Embodiments of the present invention will be described below with reference to the drawings. 1 to 6 show an embodiment of the present invention, and FIG. 1 is a block diagram showing a configuration of a speech recognition device.

This speech recognition apparatus is configured by executing a speech recognition program in a personal computer as a hardware component.

In the present embodiment, instead of directly inputting speech to the speech recognition device, digital speech data obtained by recording with a digital speech recording device independent of the speech recognition device is used. It is assumed that the voice recognition device performs the voice recognition by importing the voice recognition file as a voice file including “.

Further, in this embodiment, it is assumed that data of the digital audio recording device is stored in a flash memory card conforming to the PCMCIA standard as a file compatible with MS-DOS.

As a digital audio file, a so-called WAV file is well known. In the present embodiment, at least the "International Voice Association" (I
VA), set by Digital Speech
It has a function of reading a so-called DSS file conforming to the “Standard” (hereinafter abbreviated as DSS) standard and performing voice recognition.

A file conforming to the DSS standard includes audio data and a header section which is additional information attached to the audio data. Various information including the information shown in FIG. 2 is recorded in the header portion. FIG. 2 is a diagram showing a main part of information recorded in a header part of the DSS file.

As shown, "header block number"
Occupies one block of storage area as binary data. Contents of this header block number (content)
Contains various data depending on the manufacturer.

Further, in the item of "recording user ID"("AuthorID" in English), ID information of the recording person is recorded as ASCII data.
A recording area of six blocks is allocated.

Further, the "operator ID" (in English, "Transc
riptionist ID "), the recording person records an ID for specifying a person who transcripts the recorded content as ASCII data, and similarly allocates a recording area of 16 blocks. Have been.

Various other information is recorded in the header portion, but these are not shown.

This speech recognition apparatus, as shown in FIG.
For example, a digital voice data capturing unit 1 as a reading unit and a file capturing unit as a reading unit including a PCMCIA card interface, an operation unit 2 as a setting unit as an input device such as a keyboard and a mouse, and a voice including a CPU and a semiconductor memory. A voice recognition execution / control unit 3 as a recognition unit and a control unit, a display unit 4 as a display unit including a CRT and a video circuit for driving the CRT, a fixed disk drive (a so-called hard disk (HDD)), and the like. And a non-volatile storage unit 5 as storage means.

The digital audio data capturing section 1
It goes without saying that a recording medium such as a miniature card or a compact flash may be attached to the PCMCIA adapter for connection. Actually, for example, a digital audio recording device for recording audio data on a miniature card is already on the market.
Through the IA adapter, an audio file recorded in the miniature card can be read.

Further, the recording medium read by the digital voice data capturing unit 1 is not limited to a PCMCIA card-shaped recording medium, but may be any one that can be used by both a digital voice recording device and this voice recognition device. Various recording media can be used. For example, when various recording media such as a floppy disk and a magneto-optical disk are used, a reading device corresponding to these recording media is used as the digital audio data capturing unit 1. Will be used as

In addition, the digital voice data capturing section 1 is not limited to a means for capturing a voice file via a recording medium, but receives information from a digital voice recording apparatus by the voice recognition apparatus. As long as it is possible, for example, RS-
It is also conceivable to take in data through communication lines such as 232C standard serial communication and Ir-DA standard infrared communication.
In addition, an interface such as a USB standard or an IEEE 1394 standard may be provided.
Alternatively, the connection may be made.

The keyboard of the operation section 2 may be integrated with the main body, such as a portable personal computer or a word processor, or may be separate and connected by a cable. No problem.

The mouse of the operation unit 2 is not limited to a mouse connected to the main unit via a cable, but may be a trackball built in the main unit, a pointing device using a piezoelectric element or the like, or a display unit. Any device, such as a touch panel integrated with the display unit 4, may be used as long as it can indicate a specific location or range on the screen of the display unit 4.

Regarding the voice recognition execution / control unit 3,
In order to process the compressed audio data at high speed,
A DSP (digital signal processor) may be provided.

The display section 4 may be integral with the main body, or may be separate and connected via a cable or the like. Of course, this display section 4
The present invention is not limited to the RT, and may be an LCD or a plasma display.

The non-volatile storage section 5 is not limited to a hard disk, but may be any non-volatile and rewritable storage means such as an MO or a flash ROM. And the temporary data may be stored in a rewritable medium such as an HDD or an MO.

Further, the non-volatile storage unit 5 may be built in the main body, or may be provided with an appropriate interface such as SCSI in the main body, and may be used as an external storage device connected by a cable or the like via this interface. You may comprise.

The nonvolatile storage unit 5 stores, for example, “Windows 95” or “Windows” of Microsoft Corporation.
s98 ”or an OS (Operating System) similar thereto, a program for reading a DSS file and converting it into a data format convenient for performing voice recognition processing, a voice recognition program for performing voice recognition, Various driver programs for operating the display unit 4 and other devices such as a printer, a file for storing various setting contents, a plurality of font files, a plurality of template files, and a plurality of standard documents Files are stored.

The voice recognition program has a function for defining the association between the ID information such as the recording user ID and the operator ID and the display mode such as fonts. An ID association definition file describing the contents is created and stored in a directory which is a specific storage area of the nonvolatile storage unit 5.

Although not shown, a printer may be built in or connected to print the contents displayed on the screen, that is, the text obtained as a result of the voice recognition. good.

Next, FIG. 3 is a diagram showing an example of a display screen when the voice recognition program is started.

On the display screen 11, a pull-down menu 12 is provided in the upper part, buttons 13 composed of icons and the like are arranged below the pull-down menu 12, and a document for displaying the text of the recognition result is further displayed below the button 13. Display window 14
It has become.

The pull-down menu 12 displays each word indicating a function and each shortcut key (English character shown in parentheses) for activating the function in a horizontal line.
This is a part displaying a list of functions that can be realized by the program.

The pull-down menu 12 is shown in FIG.
In the example shown in (1), file (F), edit (E), display (V), insert (I), dictation (D), read out (R), format (F), and help (H) are arranged in order from the left side. Is displayed.

In many cases, such menus have a hierarchical structure. When any one of the items is selected and clicked with a mouse or the like, a list of lower menus is displayed immediately below the menu. ing. Such a display menu is called a pull-down menu.

For example, when "format" in the above-mentioned items is clicked on with a mouse, a list of a lower layer menu (not shown) is displayed. From this list, "ID
When an item of the "related setting menu" is selected, a screen as shown in FIG. 4 is displayed.

FIG. 4 is a diagram showing an example of a display screen when the ID-related setting menu is selected.

In the display screen 21, an "ID-related setting menu" is displayed in a portion indicated by reference numeral 22 at the top, indicating that the screen is a screen for setting an association between ID information and a font or the like. Have been.

In the display screen 21, a portion for setting ID information as indicated by reference numeral 23, a portion for defining page settings as indicated by reference numeral 24, and A part that defines the font as shown,
A part for selecting a layout or a style as shown by reference numeral 26, a part for selecting the presence or absence of a header or footer as shown by reference numeral 27, and setting a page number, and the above-mentioned respective settings as shown by reference numeral 28 And a part for confirming or canceling are arranged.

In the portion indicated by the reference numeral 23, a check box for selecting which one of the "recording user ID" and the "operator ID" is to be associated, and an input box for inputting alphanumeric characters representing the ID are provided. Is displayed.

When the operator, for example, clicks on the check box for "recording user ID" with the mouse, "recording user ID" is turned on, and a check mark is displayed in the check box as shown in the figure. Next, by clicking the input box on the right side with the mouse and then inputting alphanumeric characters with the keyboard, the input of the “recording user ID” is completed.

In the case where some associations with ID information have already been defined, a button with an inverted triangle mark on the right side of the ID input box is clicked using a mouse or the like. A list of ID information input so far is displayed immediately below the input box. That is, the ID information that has been input so far is stored and accumulated inside the nonvolatile storage unit 5 and the like, except for duplicate information.

One piece of ID information from the list displayed in this manner (in the example shown in the figure, "J" as "recording user ID")
When “007” is set), the ID information is displayed in the input box, and in the other boxes on the screen, parameters defining the association with the ID information are displayed.

The "recording user ID" is J00
In the part indicated by the reference numeral 24 for the person 7
The US letter size of 8 inches wide and 11 inches high is selected as the paper size for the page setting, and the paper is selected to be portrait.

Further, in the portion indicated by reference numeral 24, the margins are selected so as to be 1 inch each in the upper, lower, left and right directions.

Next, in the portion indicated by reference numeral 25, the font is a standard style of "Times New Roman" and 1
The one with the size of 0 points is selected.

In the portion indicated by reference numeral 26, the arrangement of the phrases is selected such that the upper left position and the indent are 1 inch on the left and right. Also, the style "Jame
"s'sformal letter" is selected. Detailed information on this style is recorded in a specific directory of the nonvolatile storage unit 5 as a template file created using another wizard. In this example, it is assumed that James's address and telephone number are arranged at predetermined positions. An overview of the state set by the arrangement of these styles and phrases can be confirmed on a small screen called a preview displayed on the right side of the portion indicated by reference numeral 26.

In the example shown in FIG. 4 in the portion indicated by the reference numeral 27, since neither the header nor the footer is marked with a check mark, both items are selected so as to be turned off. Will be. As for the page number, the setting displayed at the center of the lower part of the page is selected.

If the currently displayed state is acceptable, the OK button at the bottom left of the screen indicated by reference numeral 28 is clicked on with a mouse, and the ID information association definition file is stored in the specified directory in the nonvolatile storage unit 5. Is created. At this time, if a file with the same name already exists, the file is overwritten and the new setting becomes effective.

When the user clicks the second cancel button from the left of the portion indicated by reference numeral 28, the ID
This ID-related setting menu is ended without newly creating or overwriting the information-related definition file.

On the other hand, when the default button at the third position from the left of the portion indicated by reference numeral 28 is clicked, all the parameters as described above are reset to the default values (predetermined values) which are predetermined standard values. It is to be set.

Further, when the help button on the right end of the portion indicated by the reference numeral 28 is clicked, a supplementary explanation on the ID-related setting menu is displayed.

In the box portion for setting parameters as described above, where there is a button marked with an inverted triangle on the right side, a list of options is displayed by clicking the button with the mouse. It has become.

For example, at the item of the arrangement of the parts indicated by reference numerals 26 and 27, by clicking a button marked with an inverted triangle, options such as left alignment, center, right alignment and the like are displayed.

In the box portion in which the numbers are displayed as described above, a button with a triangle mark on the right side thereof and a button with an inverted triangle mark are arranged vertically adjacent to each other. In each of the locations, the lowest digit changes by one step each time one of the buttons is clicked with the mouse.

For example, 1.0 shown in reference numerals 24 and 26
When a button with a triangle mark is clicked in a box portion displayed as inches (1.0 "), the number is changed to 1.1, 1.2,
When the button with the inverted triangle mark is clicked from the state where 1.0 inch is displayed, the number becomes 0.9 each time the button is clicked.
It changes to 0.8 and 0.7.

By such an operation, it is possible to define the association between each parameter for setting the display mode of the text and the ID information.

FIG. 5 is a flowchart showing the overall operation of the speech recognition apparatus.

The process enclosed by a broken line in the figure is a portion operated by the operator of the voice recognition device, and the other portion is a portion that the voice recognition device automatically performs a process.

As described above, the flash memory card conforming to the PCMCIA standard contains a digital voice file recorded by a digital voice recorder or the like which is separate from the voice recognition device.
It is assumed that it has already been saved.

Then, this flash memory card is
It is installed in the slot of the digital audio data acquisition unit 1 which is a PCMCIA card interface.

When the operator wants to recognize a specific voice file by voice, the operator selects the item “dictation” from the menu on the display screen of FIG. Then
An item (not shown) such as "operator form"("TranscribeFrom" in English) is displayed in the lower-level menu list displayed in a pull-down display. Therefore, by clicking this item with a mouse, a screen (not shown) for selecting an audio file is displayed.

On this display screen, when a drive to which a flash memory card mounted in the slot of the digital audio data capturing section 1 is assigned is selected, a readable audio file, that is, the extension is DSS. A list of files will be displayed.

Then, a voice file to be subjected to voice recognition is selected from these (step S1), and an operation of opening the file is performed by clicking an "open" button with a mouse (step S2). .

As a result, the file read command is executed, and the voice recognition device accesses the selected voice file in the flash memory card,
The information recorded in the header of the audio file is read (step S3).

Subsequently, a process of automatically setting a display mode such as a font in association with the ID information read from the header of the audio file is executed (step S).
4). This process will be described later in detail.

When this process is completed, the digital audio data acquisition unit 1 sequentially reads the audio data in the audio file by a fixed amount (step S5).
This is analyzed by the voice recognition execution / control unit 3 and subjected to voice recognition to perform a conversion process into a sentence (step S6).

Then, the sentence obtained as a result of the speech recognition is
It is displayed on the document display window 14 shown in FIG. 3 of the display unit 4 such as a CRT (step S7). At this time, ID
The display is performed according to a display mode such as a font and a format set in association with the information.

Thereafter, it is determined whether or not the voice data has been completed (step S8). If voice data still remains, the process returns to step S5, where the voice is read again from the continuation of the previously read portion. A certain amount of data is read in, speech recognition is performed, and the resulting sentence is added to the previously displayed sentence in order and displayed.

This routine is repeatedly executed until all voice data has been recognized. If it is confirmed in step S8 that all of the voice data has been converted into text and displayed, this series of voice recognition processing ends.

FIG. 6 is a flowchart showing the details of the ID association automatic setting process in step S4 in FIG. The ID association automatic setting process is a process for automatically setting a font and the like in association with ID information as described above.

FIG. 6 shows an example in which, when the association has been set for both the “recording user ID” and the “operator ID”, the “recording user ID” has priority. It goes without saying that the “operator ID” can be prioritized.

First, a case will be described in which the association with the “recording user ID” has already been set.

The header of the digital audio file read from the flash memory card in step S3 is examined to determine whether or not the "recording user ID" is recorded (step S11).
If the ID information association definition file defining the association with the "recording user ID" exists in the specific directory of the non-volatile storage unit 5 if it is recorded (step S17). ), It is determined whether or not it exists (step S18).

When the corresponding ID information association definition file is found, the information recorded therein is checked one after another, and the setting of various parameters related to the text display is changed accordingly.

That is, first, it is checked whether or not the association between the font and the ID information is defined (step S20). If the font is defined, the font is changed according to the definition (step S21). (Step S22).

Next, it is checked whether or not the association with the ID information is defined for the page setting (step S).
23) If it is defined, the page setting is changed accordingly (step S24), and if it is not defined, the default page setting is made (step S25).

Subsequently, it is checked whether or not the association with the ID information is defined for the format (step S2).
6) If it is defined, the format is changed according to it (step S27), and if it is not defined, the default format is set (step S28).

Then, it is checked whether or not the association with the ID information is defined for the template indicating the template of the text (step S29). If the association is defined, the specified template file is read (step S3).
0), if not defined, the default template file is read (step S16).

In this way, various settings relating to the display mode of the sentence are automatically performed in association with the “recording user ID” recorded in the header portion of the audio file.

This setting is reflected on the text obtained by voice recognition of the voice data, and is displayed on the document display window 14 or printed.

If the recording of the "recording user ID" does not exist in the header of the audio file in step S11, it is determined whether the "operator ID" is recorded in the header of the audio file. Is detected (step S12).

When the “operator ID” is detected, the “operator ID” is stored in the specific directory of the nonvolatile storage unit 5 as in the case of the “recording user ID” as described above. A search is performed to determine whether or not an ID information association definition file defining the association exists (step S19).

If the corresponding ID information association definition file is found, the above-described processing from step S20 is performed, the information recorded therein is checked one after another, and the display mode of the text is changed accordingly. Go.

That is, similarly to the case of the above-mentioned “recording user ID”, the presence / absence of the association of the “operator ID” with the font, with the page setting, with the format setting, and with the template is checked. For items for which association is defined, the setting is changed in accordance with the information, and for items without definition, the default state is set.

As described above, various settings relating to the display mode of the text are automatically performed in association with the “operator ID” recorded in the header portion of the audio file.

Even when the "recording user ID" or the "operator ID" is recorded in the header portion of the audio file, if the association between the ID information and each parameter is not defined at all, the ID is recorded. Since the association definition file does not exist, the definition file cannot be found even if a file search is performed in steps S18 and S19.

In such a case, all the items are set to the default state as described above. That is, the font is set to the default state (step S1).
3) The page setting is set to a default state (step S14), and the format is set to a default state (step S14).
15) Then, in step S16, a default template is read.

If the "recording user ID" and the "operator ID" are not recorded in the header of the audio file in steps S11 and S12, steps S13 to S16 are performed.
Similarly, all items are set to the default state.

As described above, the ID information association automatic setting process is completed, and the process proceeds to the next step.

In FIG. 6, the font, page setting, format, template, etc. are set according to the ID information for the sake of simplicity. And footer,
If the association between ID information and fixed phrases, logos, illustrations, etc. is defined, these are automatically set to the settings corresponding to the ID information recorded in the header of the read digital audio file. It is possible to change. As an example, in the case of a Japanese letter, it is preferable to select a fixed sentence such as a greeting from a collection of sentence examples and associate it with ID information.

It is needless to say that the order in which the various parameters relating to the display mode of the sentence are automatically set is not limited to the order described above.

According to such an embodiment, it is possible to easily display a sentence automatically reflecting the preference of the recorder by utilizing the ID information recorded in the header of the audio file.

Further, since the display mode of the sentence displayed on the screen is automatically reflected in the print result, it is also possible to easily obtain a printout of the sentence automatically reflecting the preference of the recorder. It is.

In this way, when a digital voice file recorded by a digital voice recording device is converted into a text by performing voice recognition, the preference of the recorder or the operator regarding the display mode of the text can be automatically determined without complicated settings. Can be reflected.

The present invention is not limited to the above-described embodiment, and it is needless to say that various modifications and applications can be made without departing from the gist of the invention.

[Appendix] According to the above-described embodiment of the present invention, the following configuration can be obtained.

(1) File capturing means for capturing a voice file including voice data and ID information, and voice recognition for voice-recognizing voice data of the voice file captured by the file capturing means and converting it into text. Means,
Set the parameters related to the display mode of the sentence and set the above ID
Setting means for associating with the information, ID information of the voice file fetched by the file fetching means, and the sentence converted by the voice recognition means based on the parameter associated with the ID information A speech recognition apparatus, comprising: a control unit for controlling the display mode of the above;

(2) When the captured audio file does not include ID information, the control means changes the display mode of the text converted from the audio data of the audio file into a standard display mode set in advance. (1) The speech recognition device according to (1), wherein control is performed based on parameters according to (1).

(3) If there are a plurality of parameters associated with the ID information of the fetched audio file and one or more of these parameters are not set, the control means , Based on preset parameters related to the standard display mode,
The speech recognition device according to claim 1, wherein the speech recognition device controls a display mode of a sentence obtained by converting the speech data of the speech file.

(4) There is provided storage means for storing the ID information of the fetched audio file, excluding duplication.
The control means controls the display so that the ID information accumulated in the storage means is displayed in a list on the display means. The setting means selects the ID information from the list displayed on the display means. (1), (2), or (3), characterized in that it is configured to be able to set an association with a parameter related to a display mode of a sentence. Voice recognition device.

(5) The file fetching means fetches the audio file from a recording medium in which audio data and a header section including ID information are recorded as a digitized audio file. (1).

(6) The voice recognition apparatus according to (1), wherein the file capturing means captures the voice file via a communication line.

(7) The voice recognition apparatus according to the additional remark (1), wherein the display mode includes a mode relating to the font and the layout of the text that has been voice-recognized.

(8) The speech recognition apparatus according to the appendix (1), wherein the display mode includes a mode relating to a template indicating a template of a sentence.

(9) The voice recognition device according to the appendix (1), wherein the ID information includes a recording user ID for identifying a recording user.

(10) The voice recognition device according to (1), wherein the ID information includes an operator ID for identifying an operator operating the voice recognition device.

Therefore, according to the invention described in the appendix (1), it is possible to automatically reflect the preference of the recorder or the operator in the display mode of the sentence recognized by the voice, thereby improving the work efficiency. Can be achieved.

According to the invention described in Appendix (2),
The same effects as those of the invention described in Appendix (1) can be obtained, and even when the audio file does not include ID information, a standard display mode can be obtained without performing special settings.

Further, according to the invention described in the supplementary note (3), the same effects as those of the invention described in the supplementary note (1) can be obtained, and some of the plurality of parameters associated with the ID information are not set. Even in such a case, a standard display mode can be adopted for the parameter.

According to the invention described in Supplementary Note (4), the same effect as the invention described in Supplementary Note (1), (2), or (3) can be obtained, and by storing ID information, The ID information can be selected and associated with the parameter related to the display mode, and the operability is improved.

According to the invention described in the supplementary note (5), the same effects as those of the invention described in the supplementary note (1) can be obtained, and audio data can be captured using a recording medium.

According to the invention described in the supplementary note (6), the same effects as those of the invention described in the supplementary note (1) can be obtained, and audio data can be captured via a communication line.

According to the invention described in Supplementary Note (7), the same effect as the invention described in Supplementary Note (1) can be obtained, and the font and layout can be easily set to the preference of the recorder or the operator. It becomes possible.

According to the invention described in the supplementary note (8), the same effect as the invention described in the supplementary note (1) can be obtained, and a template indicating a template of a sentence can be easily set to a preference of a recorder or an operator. It is possible to do.

According to the invention described in the supplementary note (9), the same effect as that of the invention described in the supplementary note (1) can be obtained, and at least the recording user can be easily identified by the ID information.

According to the invention described in the supplementary note (10), the same effects as those of the invention described in the supplementary note (1) can be obtained, and at least the operator can be easily identified by the ID information.

[0133]

As described above, according to the speech recognition apparatus of the first aspect of the present invention, the display mode of a sentence that has undergone voice recognition is controlled based on the additional information. It is possible to automatically reflect the preference of the operator, and work efficiency can be improved.

Further, according to the speech recognition apparatus of the present invention according to the second aspect, the same effects as those of the first aspect can be obtained, and even when there is no additional information or when the additional information is invalid, the special effect can be obtained. A predetermined display mode can be obtained without making any necessary settings.

Further, according to the speech recognition apparatus of the present invention according to the third aspect, the same effect as that of the first aspect can be obtained, and the parameters can be set by the setting means. The sentence can be controlled to a desired display mode.

According to the speech recognition apparatus of the present invention according to the fourth aspect, the same effect as that of the third aspect can be obtained, and even if there is a parameter not associated with the additional information, the parameter Is set to a predetermined value, so that the display mode can be appropriately controlled.

In addition, according to the speech recognition apparatus of the present invention, the same effects as those of the first aspect can be obtained, and the display mode of the text can be controlled based on the user ID information. it can.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a speech recognition device according to an embodiment of the present invention.

FIG. 2 is a diagram showing a main part of information recorded in a header part of a DSS file in the embodiment.

FIG. 3 is a diagram showing an example of a display screen when a voice recognition program is started in the embodiment.

FIG. 4 is a diagram showing an example of a display screen when an ID-related setting menu is selected in the embodiment.

FIG. 5 is a flowchart showing an overall operation of the speech recognition device of the embodiment.

FIG. 6 is a flowchart showing details of an ID association automatic setting process in step S4 of FIG. 5;

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Digital voice data taking-in part (reading means, file taking-in means) 2 ... Operation part (setting means) 3 ... Voice recognition execution and control part (speech recognition means, control means) 4 ... Display part (display means) 5 ... Non-volatile storage unit (storage means)

Claims (5)

[Claims] 1. A voice recognition apparatus for voice-recognizing voice data and additional information relating to the voice data added to the voice data and converting the voice data into a sentence, wherein: a reading unit for reading the additional information; And a control means for controlling a display mode of the converted sentence. 2. The method according to claim 1, wherein the control unit does not have the additional information.
2. The speech recognition device according to claim 1, wherein, when invalid, the display mode of the converted text is controlled based on information on a predetermined display mode. 3. 3. The control means includes a setting means for setting a parameter relating to a display mode of a sentence based on the additional information, and controls a display mode of the converted sentence based on the parameter. The voice recognition device according to claim 1. 4. When setting one or more parameters based on the additional information, the setting means determines in advance, among parameters to be set, parameters not associated with the additional information. 4. The speech recognition device according to claim 3, wherein the value is set to a predetermined value. 5. The speech recognition apparatus according to claim 1, wherein the control means controls a display mode of the converted text based on user ID information among the additional information.