JP2017181727A - Voice translating device, voice translating method, and voice translating program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make possible improvement of the accuracy of speech analysis of inputted voice and enhanced efficiency of the workload of analytical work thereby to improve the accuracy of voice translation.SOLUTION: A voice translating device according to an aspect is equipped with an input unit for accepting input of the user's voice, a translating unit that translates the contents of inputted voice into another language, an output unit for outputting the translated contents of the inputted voice in voice and/or written text, a log recording unit for recording a log of the inputted voice, and a mode switching unit for switching over between a first mode for voice inputting of conversation between the user and a user's conversation partner and a second mode for voice inputting in non-conversational actions between the user and the partner. The log recording unit distinguishes the log of inputted voice in the first mode and the log of inputted voice in the second mode in recording them.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a speech translation device, a speech translation method, and a speech translation program.

  To make conversations between people who do not understand each other's language, for example, conversations between shop assistants (sales staff at restaurants, etc.) and foreigners (tourists from abroad, etc.) A speech translation technology has been proposed in which the text content is machine-translated into the language of the other party and displayed on the screen, or the text content is played back using speech synthesis technology (for example, Patent Document 1). In addition, a speech translation application that operates on an information terminal such as a smartphone that embodies such speech translation technology has been put into practical use (see, for example, Non-Patent Document 1).

Japanese Patent Laid-Open No. 9-34895

U-STAR Consortium homepage [Search on February 1, 2016], Internet <URL: http://www.ustar-consortium.com/app_en/app.html>

  In such a conventional speech translation application or the like, the content of the input speech of the speaker is recorded in time series together with the utterance date and time (hereinafter, the recorded data is referred to as “log of input speech”), and the conversation. The analysis (utterance analysis) of the utterance content, the utterance form, the utterance situation, the utterance tendency, the utterance frequency, etc. may be performed. The result of this utterance analysis is contributed to learning of a speech recognition model (acoustic model, language model, etc.) used in a speech translation application or the like, and the accuracy of speech translation is improved.

By the way, the purpose of using the speech translation application is to use the actual conversation (communication) or to confirm in advance the content to be translated in a situation where there is no conversation partner (interactive person). In addition, for example, use of the following forms is conceivable.
(1) Trial of speech translation application or trial to get used to its operation (2) Dictionary lookup to examine the meaning of words and idioms (3) Translation of sentences other than conversation (articles, lyrics, quotes, dialogues, etc.) (4) Practice pronunciation of foreign languages

  However, not only the log of the input voice in actual conversation (actual use) but also the log of the input voice at the time of use in the form as shown in the above (1) to (4) is recorded. It is not desirable to perform utterance analysis of input speech during actual use from the viewpoint of improving analysis accuracy, improving efficiency of analysis work, and improving translation accuracy.

  Therefore, the present invention has been made in view of such circumstances, and can improve the accuracy of speech analysis of input speech and improve efficiency by reducing the load of analysis work, thereby improving the accuracy of speech translation. An object of the present invention is to provide a speech translation apparatus, a speech translation method, and a speech translation program.

  In order to solve the above problems, a speech translation apparatus according to an aspect of the present invention includes an input unit for inputting a user's speech, a translation unit that translates the content of the input speech into content of a different language, and translation of the input speech An output unit is provided for outputting the contents (parallel translation) as audio and / or text. The speech translation apparatus includes: a log recording unit that records a log of input speech; a first mode for performing speech input in a conversation between a user and a conversation person (conversation partner); a user and a conversation person; And a mode switching unit that switches between the second mode for performing voice input in an operation other than the conversation of the user, and the log recording unit includes a log of input voice in the first mode and a second mode. Input audio logs are recorded separately from each other. It should be noted that “conversation” includes use for preparing for a conversation such as confirming in advance the contents to be translated in a situation where there is no interlocutor.

  Specifically, the mode switching unit may display switching means for the user to switch between the first mode and the second mode on the user device used by the user.

  Further, the log recording unit may record the input voice log in the first mode and the input voice log in the second mode in the same database.

  Furthermore, it is preferable to further include an analysis unit that performs speech analysis using only the log of the input speech in the first mode.

  A speech translation method according to an aspect of the present invention is a method using a speech translation apparatus including an input unit, a translation unit, an output unit, a log recording unit, and a mode switching unit. That is, in the method, the step of inputting the user's voice, the step of the translation unit translating the content of the input voice into the content of a different language, and the step of outputting the translation content by voice and / or text A step in which the log recording unit records a log of the input voice, a mode switching unit performs a voice input in a conversation between the user and the conversation person, and a conversation other than the conversation between the user and the conversation person. And switching to the second mode for performing voice input in the operation. In the step of recording the log, the input voice log in the first mode and the input voice log in the second mode are recorded separately from each other.

  A speech translation program according to an aspect of the present invention includes a computer (not limited to a single type or a single type, but may be a plurality or a plurality of types; the same applies hereinafter), an input unit for inputting a user's voice, A translation unit that translates content into different language content, an output unit that outputs the translated content in speech and / or text, a log recording unit that records a log of input speech, and voice input in a conversation between a user and a conversation person And a mode switching unit that switches between a first mode for performing voice input and a second mode for performing voice input in operations other than conversation between the user and the interlocutor. Then, the log recording unit records the input voice log in the first mode and the input voice log in the second mode separately from each other.

  According to the present invention, the first mode for performing speech input in the conversation between the user and the conversation person (that is, when the speech translation apparatus or the like is actually used) and the operation other than the conversation between the user and the conversation person (that is, trial) , Dictionary lookup, sentence translation other than conversation, pronunciation practice, etc.) are provided for the second mode, and these modes can be switched as appropriate. Also, the input voice log in the first mode (actual use) and the input voice log in the second mode (use other than actual use) can be recorded separately from each other. Therefore, only the log of the input voice in the first mode (that is, only the utterance information at the time of actual use) can be easily extracted, and by using it, the utterance analysis at the time of actual use can be accurately performed, and The load can be reduced. As a result, it is possible to improve the accuracy of speech analysis of input speech and increase the efficiency of analysis work, thereby improving the accuracy of speech translation.

1 is a system block diagram schematically showing a preferred embodiment of a network configuration and the like related to a speech translation apparatus according to the present invention. It is a flowchart which shows an example of the flow (part) of the process in suitable one Embodiment of the speech translation apparatus by this invention. (A) thru | or (C) are top views which show an example of the transition of the display screen in the information terminal by 1st Embodiment. It is a schematic diagram which shows an example of the data structure of an audio log database.

  Hereinafter, embodiments of the present invention will be described in detail. The following embodiments are examples for explaining the present invention, and are not intended to limit the present invention only to the embodiments. The present invention can be variously modified without departing from the gist thereof. Furthermore, those skilled in the art can employ embodiments in which the elements described below are replaced with equivalent ones, and such embodiments are also included in the scope of the present invention. Furthermore, positional relationships such as up, down, left, and right shown as needed are based on the display shown unless otherwise specified. Furthermore, various dimensional ratios in the drawings are not limited to the illustrated ratios.

(Device configuration)
FIG. 1 is a system block diagram schematically showing a preferred embodiment such as a network configuration related to a speech translation apparatus according to the present invention. In this example, the speech translation apparatus 100 includes a server 20 that is electronically connected to the information terminal 10 (user apparatus) used by the user via the network N (but is not limited to this).

  The information terminal 10 employs a user interface such as a touch panel and a display with high visibility, for example. The information terminal 10 here is a portable tablet terminal device including a mobile phone represented by a smartphone having a communication function with the network N. The information terminal 10 further includes a processor 11, a storage resource 12, a voice input / output device 13, a communication interface 14, an input device 15, a display device 16, and a camera 17. In addition, the information terminal 10 operates by the installed speech translation application software (at least a part of the speech translation program according to the embodiment of the present invention), so that a part of the speech translation apparatus according to the embodiment of the present invention or It functions as a whole.

  The processor 11 includes an arithmetic logic unit and various registers (program counter, data register, instruction register, general-purpose register, etc.). Further, the processor 11 interprets and executes speech translation application software, which is the program P10 stored in the storage resource 12, and performs various processes. The speech translation application software as the program P10 can be distributed from the server 20 through the network N, for example, and may be installed and updated manually or automatically.

  The network N includes, for example, a wired network (a short-range communication network (LAN), a wide-area communication network (WAN), a value-added communication network (VAN), etc.) and a wireless network (mobile communication network, satellite communication network, Bluetooth ( Bluetooth (registered trademark)), WiFi (Wireless Fidelity), HSDPA (High Speed Downlink Packet Access), etc.).

  The storage resource 12 is a logical device provided by a storage area of a physical device (for example, a computer-readable recording medium such as a semiconductor memory), and an operating system program, a driver program, various data, etc. used for processing of the information terminal 10 Is stored. Examples of the driver program include an input / output device driver program for controlling the audio input / output device 13, an input device driver program for controlling the input device 15, and a display device driver program for controlling the display device 16. Can be mentioned. Furthermore, the voice input / output device 13 is, for example, a general microphone and a sound player capable of reproducing sound data.

  The communication interface 14 provides a connection interface with the server 20, for example, and is configured from a wireless communication interface and / or a wired communication interface. The input device 15 provides an interface for accepting an input operation by a tap operation such as an icon, a button, or a virtual keyboard displayed on the display device 16, and is externally attached to the information terminal 10 in addition to the touch panel. Various input devices can be exemplified.

  The display device 16 provides various types of information as an image display interface to a user or a dialoguer (conversation partner) as necessary. Examples thereof include an organic EL display, a liquid crystal display, and a CRT display. The camera 17 is for capturing still images and moving images of various subjects.

  The server 20 is constituted by, for example, a host computer having a high arithmetic processing capability, and expresses a server function by operating a predetermined server program in the host computer, for example, a speech recognition server, a translation server, And a single or a plurality of host computers functioning as a speech synthesis server (in the drawing, it is indicated by a single, but is not limited thereto). Each server 20 includes a processor 21, a communication interface 22, and a storage resource 23.

  The processor 21 is composed of an arithmetic and logic unit for processing arithmetic operations, logical operations, bit operations and the like and various registers (program counter, data register, instruction register, general-purpose register, etc.), and is stored in the storage resource 23. P20 is interpreted and executed, and a predetermined calculation processing result is output. The communication interface 22 is a hardware module for connecting to the information terminal 10 via the network N. For example, the communication interface 22 is a modulation / demodulation device such as an ISDN modem, an ADSL modem, a cable modem, an optical modem, or a soft modem.

  The storage resource 23 is a logical device provided by, for example, a storage area of a physical device (a computer-readable recording medium such as a disk drive or a semiconductor memory), and each includes one or a plurality of programs P20, various modules L20, and various databases. D20 and various models M20 are stored. The storage resource 23 also stores a plurality of question phrases prepared in advance for the user to speak to the conversation person, input voice history data, various setting data, and the like.

  The program P20 is the above-described server program that is the main program of the server 20. In addition, the various modules L20 perform a series of information processing related to requests and information transmitted from the information terminal 10, so that they are appropriately called and executed during the operation of the program P10 (moduleized subprograms). ). Examples of the module L20 include a speech recognition module, a translation module, and a speech synthesis module.

  The various databases D20 include various corpora required for speech translation processing (for example, in the case of Japanese and English speech translation, a Japanese speech corpus, an English speech corpus, a Japanese character (vocabulary) corpus, an English character) (Vocabulary) Corpus, Japanese dictionary, English dictionary, Japanese-English bilingual dictionary, Japanese-English bilingual corpus, etc.), voice database, management database for managing information about users, and the like. Examples of the various models M20 include acoustic models and language models used for speech recognition.

  An example of operations and operations of speech translation processing in the speech translation apparatus 100 configured as described above will be further described below.

  FIG. 2 is a flowchart showing an example of (a part of) the processing flow in the speech translation apparatus 100. 3A to 3C are plan views showing an example of (a part of) a transition in the screen display of the information terminal 10. Here, the user of the information terminal 10 is a store clerk of a restaurant or the like that speaks Japanese, and the foreigner who speaks English, Chinese (for example, two types depending on the typeface), or Korean Is assumed (however, the language is not limited to this).

  First, when the user (store clerk) starts the application (step SU1), the dialog language selection screen shown in FIG. 3A is displayed on the display device 16 of the information terminal 10 (step SJ1). The language selection screen includes Japanese text T1 for prompting the user to ask the language about the conversation person, English text T2 for asking the conversation person about the language, and a plurality of typical languages (here However, the language button 31 indicating English, Chinese (for example, two types depending on the typeface), or Hangul) is displayed.

  At this time, as shown in FIG. 3A, the Japanese text T1 and the English text T2 are separated by different areas on the screen of the display device 16 of the information terminal 10 by the processor 11 and the display device 16, and They are displayed in reverse directions (different directions; upside down in the figure). Thereby, when a conversation is performed in a state where the user and the conversation person face each other, the user can easily confirm the Japanese text T1, while the conversation person can easily confirm the English text T2. Further, since the Japanese text T1 and the English text T2 are displayed separately, there is an advantage that they can be clearly distinguished and further confirmed.

  The user presents the display of the English text T2 on the language selection screen to the dialog person and asks the dialog person to tap, for example, an English button, or the user himself selects the language of the dialog person. be able to. When the language of the conversation person is selected in this manner, the processor 21 of the server 20 and the processor 11 of the information terminal 10 display a standby screen for voice input in Japanese and English on the display device 16 as the home screen (FIG. 3). (B); Step SJ2). On this standby screen, the Japanese text T3 asking which of the user's language and the talker's language is to be spoken, the Japanese input button 32a for performing Japanese speech input, and the English speech input. An English input button 32b is displayed.

  Also, on this standby screen, a voice button 33 for selecting a list display of a plurality of preset question phrases, and the language selection screen of FIG. A language selection button 34 for re-selecting the language, a history button 35 for selecting the history display of the voice input contents made so far, and a setting button 36 for performing various settings of the application software are also displayed. The

  Here, in FIG. 2, as a subsequent process, the user selects the setting button 36, and the case is focused on whether or not a “practice mode” (second mode) to be described later is turned on ( The flow of steps SU2, SU3) is shown. Note that in normal speech translation processing, speech input for conversation and / or conversation preparation between the user and the conversation person can be performed from the standby screen shown in FIG. Mode "(referred to as the first mode). Hereinafter, the flow of processing in the normal mode and the practice mode will be described in order.

[Normal mode (first mode)]
When the user taps the Japanese input button 32a and selects Japanese voice input on the standby screen shown in FIG. 3B, the voice input is enabled. In this state, when the user utters a matter to be communicated to the interlocutor, voice input is performed through the voice input / output device 13 (step SJ3). The processor 11 of the information terminal 10 generates an audio signal based on the audio input, and transmits the audio signal to the server 20 through the communication interface 14 and the network N. As described above, the information terminal 10 itself, or the processor 11 and the voice input / output device 13 function as an “input unit”.

  The processor 21 of the server 20 receives the voice signal through the communication interface 22 and performs voice recognition processing (step SJ4). At this time, the processor 21 calls the necessary module L20, database D20, and model M20 (speech recognition module, Japanese speech corpus, acoustic model, language model, etc.) from the storage resource 23, and obtains “sound” of the input speech. Convert to "reading" (character). As described above, the processor 21 or the server 20 functions as a “voice recognition server” as a whole.

  Next, when the input voice is recognized, the processor 21 determines whether the voice input is performed in the “normal mode” or the “practice mode” (mode determination). Here, since the voice input is in the normal mode, for example, “1” is assigned as a flag to the recognized input voice (step SJ5). Then, the processor 21 records the log of the recognized input voice in the voice log database D21 which is one of the databases D20 stored in the storage resource 23 (step SJ6). As described above, the processor 21 and the storage resource 23 function as a “log recording unit”.

  Here, FIG. 4 is a schematic diagram showing an example of the data structure of the voice log database D21. The voice log database D21 can be expressed as, for example, a data table 40. The data table 40 is a data number 41, a mode type 42, a user ID 43, a recording date (utterance date) 44, an input for each voice log data D. It has a plurality of data records such as voice text 45. Among these data records, the above-described flag is recorded in the mode type 42. In this normal mode, the processor 21 associates the recognized input voice log with the voice log data D with the data number 41 = “1” or “3”, for example, and sets the mode type 42 to the flag assigned to the normal mode. Is recorded in the voice log database D21 as “1”.

  Next, the processor 21 proceeds to multilingual translation processing for translating the recognized “reading” (characters) of the recognized speech into another language (step SJ7). At this time, the processor 21 calls the necessary module L20 and database D20 (translation module, Japanese character corpus, Japanese dictionary, English dictionary, Japanese-English bilingual dictionary, Japanese-English bilingual corpus, etc.) from the storage resource 23 and recognizes them. The resulting input speech “reading” (character string) is properly sorted and converted into Japanese phrases, clauses, sentences, etc., the English corresponding to the conversion result is extracted, and these are sorted according to the English grammar. To natural English phrases, clauses, sentences, etc. As described above, the processor 21 also functions as a “translation unit”, and the server 20 also functions as a “translation server” as a whole. If the input voice is not recognized well, the voice can be re-input (screen display is not shown).

  Further, the processor 21 stores the content of the recognized input voice in the storage resource 23. Next, when the multilingual translation processing and the storage processing of the contents of the input speech are completed, the processor 21 proceeds to speech synthesis processing (step SJ8). At this time, the processor 21 calls the necessary module L20, database D20, and model M20 (speech synthesis module, English speech corpus, acoustic model, language model, etc.) from the storage resource 23, and the English phrase that is the translation result, Convert clauses, sentences, etc. to natural speech. As described above, the processor 21 also functions as a “speech synthesizer”, and the server 20 also functions as a “speech synthesizer” as a whole.

  Next, the processor 21 generates a voice signal for voice output based on the synthesized voice, and transmits the voice signal to the information terminal 10 through the communication interface 22 and the network N. The processor 11 of the information terminal 10 receives the audio signal through the communication interface 14, and performs an audio output process using the audio input / output device 13 (step SJ9). As described above, the processor 11 and the voice input / output device 13 function as an “output unit”.

[Practice mode (second mode)]
On the other hand, when the user taps the setting button 36 on the standby screen shown in FIG. 3B (“Yes” in step SU2), the processor 11 of the information terminal 10 issues a command for displaying a list of various setting menus. A signal is transmitted to the server 20. The processor 21 of the server 20 that has received the command signal transmits display image data of the setting menu list to the processor 11 of the information terminal 10, and the processor 11 displays a setting menu screen shown in FIG. (Step SJ10). On this setting menu screen, a Japanese setting menu 37 that can be understood by the user is displayed as a list in horizontal writing in the vertical direction, and the user selects a desired setting menu by tapping the text portion of each setting menu. Can do.

  In the setting menu 37, a “practice mode” is prepared, and a slide switch 38 is displayed adjacent to the Japanese text T4 representing the practice mode. The slide switch 38 is used to switch the practice mode ON / OFF. When the user turns on the slide switch 38 as necessary, the slide switch 38 is used to get used to the trial or operation of the speech translation application. “Practice” as a trial can be performed separately from the normal mode. As described above, the slide switch 38 corresponds to a “switching unit” for switching between the first mode and the second mode, and the processors 11 and 21 function as a “mode switching unit”.

  When the user turns on the practice mode and closes the setting menu, the standby screen shown in FIG. 3B is displayed again on the display device 16, and steps SJ3 to SJ in the “normal mode” described above are performed. The same procedure can be executed. At this time, since the process in the practice mode is different from the process in the normal mode only in the processes in steps SJ5 and SJ6, detailed description here is omitted for the processes in the other steps.

  That is, in this practice mode, the processor 21 of the server 20 assigns, for example, “2” as a flag to the recognized input voice in step SJ5. Then, in this practice mode, in step SJ6, the processor 21 associates the recognized input voice log with the voice log data D with the data number 41 = "2", for example, and assigns the mode type 42 to the practice mode. Is recorded in the voice log database D21 as “2” which is a flag.

[Speech analysis]
As described above, in the input voice log recording in step SJ6, the input voice log in the normal mode (first mode) and the input voice log in the practice mode (second mode) are distinguished from each other by flags. Recorded in the voice log database D21. The processor 21 of the server 20 then only logs the input voice in the normal mode (first mode) among the plurality of voice log data D recorded in the voice log database D21 (that is, the mode type 42 is “1”). A certain voice log data D) is extracted, and utterance analysis is performed using it (step SJ20). As described above, the processor 21 also functions as an “analysis unit”.

  According to the speech translation apparatus 100 configured as described above, the speech translation method using the speech translation apparatus, and the speech translation program, it is possible to perform speech input in a conversation between the user and the conversation person (when the speech translation apparatus 100 is actually used). A normal mode (first mode) and a practice mode (second mode) for performing speech input in the speech translation apparatus 100, which is an operation other than the conversation between the user and the interlocutor. These modes can be switched as appropriate to use speech translation. In addition, the input voice logs in the normal mode and the practice mode are recorded in the voice log database D21 separately from each other by the respective flags, and the speech analysis can be performed using only the input voice logs in the normal mode.

  Therefore, it is possible to reliably prevent the content of the input speech uttered in the practice mode (speech content that is not necessarily uttered during actual use) from being mixed into the utterance analysis. As a result, it is possible to perform utterance analysis using only utterance information during actual use, and it is possible to reduce the amount of analysis work and reduce the load of utterance analysis. Thereby, the accuracy (learning accuracy) of utterance analysis during actual use can be improved and the efficiency of the analysis work can be improved, so that the accuracy of speech translation itself can be increased.

  Note that, as described above, each of the above embodiments is an example for explaining the present invention, and is not intended to limit the present invention to the embodiment. The present invention can be variously modified without departing from the gist thereof. For example, those skilled in the art can replace the resources (hardware resources or software resources) described in the embodiments with equivalents, and such replacements are also included in the scope of the present invention.

  In addition, as in the above embodiment, a flag may be set for both the normal mode and the practice mode, but a flag may be set for only one of them. In such a configuration, each of the modes recognized in both modes may be set. Input audio logs can be distinguished from each other. Furthermore, the log of each input voice in the normal mode and the practice mode may be recorded in separate voice log databases. Furthermore, the practice mode may be selected from a menu other than the setting menu or without going through the setting menu. For example, you may make it display the button etc. which can select practice mode directly on the screen shown to FIG. 3 (A) or FIG. 3 (B).

  Further, the steps SJ5 and SJ6 may be executed at any time as long as they are after step SJ4. Furthermore, the process of recording the input voice log in the voice log database D21 may be performed for each utterance, or a plurality of input voice logs may be recorded at a predetermined time. Also, in the practice mode, not only the trial of the speech translation application or the trial to get used to its operation, but also, for example, a dictionary lookup to examine the meaning of words and idioms, sentences other than conversation (articles, lyrics, quotes, dialogue) Etc.) can also be used for operations other than actual use, such as translation practice and pronunciation practice of foreign languages. In this way, it is possible to distinguish the input voice in the operation other than the actual use from the input voice in the actual use.

  Moreover, although the example which performs each process of speech recognition, translation, and a speech synthesis by the server 20 was described, you may comprise so that these processes may be performed in the information terminal 10. FIG. In this case, the module L20 used for these processes may be stored in the storage resource 12 of the information terminal 10 or may be stored in the storage resource 23 of the server 20. Furthermore, the database D20 that is a voice database and / or a model M20 such as an acoustic model may be stored in the storage resource 12 of the information terminal 10 or may be stored in the storage resource 23 of the server 20. . As described above, the speech translation apparatus may not include the network N and the server 20.

  Of course, a gateway server for converting a communication protocol between the information terminal 10 and the network N may be interposed. The information terminal 10 is not limited to a portable device, and may be a desktop personal computer, a notebook personal computer, a tablet personal computer, a laptop personal computer, or the like.

  According to the present invention, it is possible to improve the accuracy of speech analysis of input speech and increase the efficiency of the analysis work, thereby improving the accuracy of speech translation. The present invention can be widely used for activities such as design, manufacture, provision, and sales of programs, devices, systems, and methods in the field of providing services related to conversation between people.

10 information terminal 11 processor 12 storage resource 13 voice input / output device 14 communication interface 15 input device 16 display device 17 camera 20 server 21 processor 22 communication interface 23 storage resource 31 language button 32a Japanese input button 32b English input button 33 Button 34 Language selection button 35 History button 36 Setting button 37 Setting menu 38 Slide switch 40 Data table 41 Data number 42 Mode type 43 User ID
44 recording date 45 input speech text 100 speech translation device D speech log data D20 database D21 speech log database L20 module M20 model N network P10 program P20 program T1 Japanese text T2 English text T3 Japanese text T4 Japanese text

Then the user closes the configuration menu ON the practice mode, the display device 16 to redisplay the standby screen shown in FIG. 3 (B), the step SJ3～SJ 9 in the "normal mode" as described above Can be executed in the same procedure. At this time, since the process in the practice mode is different from the process in the normal mode only in the processes in steps SJ5 and SJ6, detailed description here is omitted for the processes in the other steps.

Claims (6)

A speech translation apparatus comprising: an input unit for inputting a user's speech; a translation unit that translates the content of the input speech into content of a different language; and an output unit that outputs the translation content of the input speech as speech and / or text Because
A log recording unit for recording the log of the input voice;
Mode switching for switching between a first mode for performing voice input in a conversation between the user and the talker and a second mode for performing voice input in an operation other than the conversation between the user and the talker And
Further comprising
The log recording unit records the input audio log in the first mode and the input audio log in the second mode separately from each other;
Speech translation device. The mode switching unit displays switching means for the user to switch between the first mode and the second mode on a user device used by the user.
The speech translation apparatus according to claim 1. The log recording unit records the input voice log in the first mode and the input voice log in the second mode in the same database;
The speech translation apparatus according to claim 1 or 2. An analysis unit that performs speech analysis using only the log of the input voice in the first mode;
The speech translation apparatus according to any one of claims 1 to 3. Using a speech translation device including an input unit, a translation unit, an output unit, a log recording unit, and a mode switching unit,
The input unit inputs a user's voice;
The translation unit translating the content of the input speech into content of a different language;
The output unit outputting the translated content in speech and / or text;
The log recording unit recording a log of the input voice;
A first mode for performing voice input in a conversation between the user and the conversation person, and a second mode for performing voice input in an operation other than the conversation between the user and the conversation person. A step of switching between and
Including
In the step of recording the log, the input voice log in the first mode and the input voice log in the second mode are recorded separately from each other,
Speech translation method. Computer
An input unit for inputting the user's voice;
A translation unit that translates the content of the input speech into content of a different language;
An output unit that outputs the translated content in audio and / or text;
A log recording unit for recording the log of the input voice;
A mode switching unit that switches between a first mode for performing voice input in a conversation between the user and the talker and a second mode for performing voice input in an operation other than the conversation between the user and the talker. When,
To function,
The log recording unit records the input audio log in the first mode and the input audio log in the second mode separately from each other;
Speech translation program.

Images