JP6511189B1 - Information processing system and transcription method - Google Patents

Abstract

An object of the present invention is to increase the secrecy of the content of a voice to be transcribed. A management device (12) divides target voice data in which a target voice of transcription is recorded into a plurality of section voice data relating to a plurality of sections. The management device 12 provides the worker device 16 with at least one of a plurality of section voice data and a set of dummy voice data. The management device 12 receives text data transcribed based on at least one of the plurality of section voice data and text data transcribed according to dummy voice data. The management device 12 generates text data in which the target speech is transcribed, using text data transcribed from at least one of the plurality of section voice data among the received text data. [Selected figure] Figure 1

Description

  The present invention relates to data processing technology, and more particularly to an information processing system and transcription method.

  There has been proposed a transcription system for generating characters from a conversation recorded as speech (see, for example, Patent Document 1). In the transcription system of Patent Document 1, the server divides voice data in which a conversation has been recorded into voice data relating to a plurality of voice segments, and transmits voice data of each voice segment to a plurality of information terminals. Each information terminal outputs a character string generated by transcribing voice data to the server, and the server combines individual character strings to construct sentence data in which the entire conversation of the original voice data is sentenced.

JP, 2008-107624, A

  It may not be desirable for the sound to be transcribed to be transmitted as it is to the operator performing the transcription, such as when confidential matters are included. The inventor considered that there is room for improvement to enhance the secrecy of the content of the speech to be transcribed.

  The present invention has been made on the basis of the above-mentioned problem recognition of the inventor, and one object is to enhance the secrecy of the content of speech to be transcribed.

  In order to solve the above problems, an information processing system according to an aspect of the present invention includes a first storage unit storing target voice data in which a target voice of transcription is recorded, and a dummy voice in which a dummy voice is recorded A second storage unit for storing data, a division unit for dividing target voice data into a plurality of section voice data relating to a plurality of sections, and a set of at least one of a plurality of section voice data and dummy voice data to an external device A receiving unit for receiving, text data obtained by transcribing based on at least one of a plurality of section voice data, and text receiving unit for receiving text data obtained by transcribing based on dummy voice data; Of the data, text obtained by transcribing the target speech using text data transcribed based on at least one of the plurality of section speech data Comprising a generation unit for generating data.

  Another aspect of the present invention is a transcription method. According to this method, an information processing system for storing target voice data in which a target voice of transcription is recorded and dummy voice data in which a dummy voice is recorded includes a plurality of sections related to a plurality of sections of target voice data. The step of dividing into voice data, the step of providing at least one of a plurality of section voice data and a dummy voice data to an external device, text data transcribed from at least one of the plurality of section voice data A step of receiving both text data transcribed based on dummy speech data, and text data transcribed based on at least one of a plurality of section speech data among the accepted text data; Generating text data that transcribes the speech of

  Note that any combination of the above-described components, and one obtained by converting the expression of the present invention between a device, a computer program, a recording medium storing a computer program, and the like are also effective as an aspect of the present invention.

  According to the present invention, the secrecy of the content of the speech to be transcribed can be enhanced.

It is a figure showing composition of a transcription system of an example. It is a block diagram which shows the functional block of the management apparatus of FIG. It is a figure which shows the example of division | segmentation of audio | voice data. It is a figure which shows the example of audio | speech data. It is a figure which shows the example of allocation of area audio | speech data. It is a figure which shows the example of the work result by a worker. It is a figure which shows the example of division | segmentation of audio | voice data.

  The transcription system according to the embodiment is a system for transcribing a combination of at least a part of speech to be transcribed (a speech relating to the user and containing secret information) and a dummy speech. Provide and cause the operator to transcribe both of those sounds. As a result, it is possible to reduce the risk of leakage of the entire content of the speech to be transcribed, and to improve the secrecy of the content of the speech to be transcribed.

  FIG. 1 shows the configuration of a transcription system 10 of the embodiment. The transcription system 10 is an information processing system that supports transcription, and includes a management device 12, a plurality of user terminals 14, and a plurality of worker devices 16. Each device of the transcription system 10 is connected via a communication network 18 including LAN, WAN, Internet, and the like. Transcription is the conversion of audio content into text, and may be called a transcription.

  The management device 12 is an information processing device that provides a web service of transcription to a plurality of user terminals 14. Detailed functions of the management device 12 will be described later.

  The plurality of user terminals 14 are information processing devices operated by a user who uses the transcription service. The plurality of user terminals 14 are operated by a user terminal 14a operated by a user a belonging to company A, a user terminal 14b operated by a user b belonging to company B, and a user c belonging to company C It includes a user terminal 14c. The user terminal 14 may be a PC, a tablet terminal, or a smartphone.

  The plurality of worker devices 16 are information processing devices that mainly perform transcription. In an embodiment, a human listens to speech and converts the speech into text. The plurality of worker devices 16 include a worker device 16x operated by the worker x, a worker device 16y operated by the worker y, and a worker device 16z operated by the worker z. The worker device 16 may be a PC, a tablet terminal, or a smartphone.

  FIG. 2 is a block diagram showing functional blocks of the management device 12 of FIG. Each block shown in the block diagram of the present specification can be realized by hardware as a processor of a computer, a CPU, an element or electronic circuit such as a memory or an electronic circuit, or a mechanical device, and software can be realized by a computer program etc. However, here we describe the functional blocks that are realized by their coordination. Therefore, it is understood by those skilled in the art that these functional blocks can be realized in various forms by a combination of hardware and software.

  The management device 12 includes a control unit 20, a storage unit 22, and a communication unit 24. The control unit 20 executes various data processing for providing a transcription service. The storage unit 22 stores data referred to or updated by the control unit 20. The communication unit 24 communicates with an external device according to a predetermined communication protocol. The control unit 20 transmits and receives data to and from the user terminal 14 and the worker device 16 via the communication unit 24.

  The storage unit 22 includes a target voice storage unit 30, a dummy voice storage unit 32, an assignment rule storage unit 34, a distribution data storage unit 36, a work result storage unit 38, a sentence storage unit 40, a correct answer storage unit 42, and an evaluation storage unit 44. Including. Note that at least a portion of the data stored in storage unit 22 may be stored in a storage device (not shown) different from management device 12, and management device 12 stores data stored in an external storage device. Of course, you may refer to and update.

  The target voice storage unit 30 stores target voice data which is voice data received from the user terminal 14 and on which a voice to be transcribed (hereinafter also referred to as “target voice”) is recorded. The dummy voice storage unit 32 stores not dummy voice data received from the user terminal 14 but dummy voice data in which dummy voice (hereinafter also referred to as “dummy voice”) is recorded. The dummy sound is a sound of content predetermined by the manager of the management apparatus 12 (for example, a person in charge of a company providing a transcription service).

  The correct answer storage unit 42 stores text data indicating the content of the dummy speech. In the embodiment, the dummy sound data is divided in advance into sound data for each section related to a plurality of sections (hereinafter also referred to as “section sound data”), and the dummy sound storage unit 32 is plurally based on the dummy sound data. Section voice data is stored. In addition, the correct answer storage unit 42 stores text data (hereinafter, also referred to as “correct answer data”) indicating the contents of each of a plurality of section voice data based on dummy voice data.

  The assignment rule storage unit 34 stores section voice data into which target voice data is divided, and a rule for assigning section voice data into which dummy voice data is divided to a worker (hereinafter also referred to as “assignment rule”). . The assignment rule will be described later in relation to the configuration of the assignment unit 56.

  The delivery data storage unit 36 is data to be delivered to each of a plurality of workers, and stores delivery data including one or more section voice data. For example, the delivery data storage unit 36 includes delivery data to the worker x (worker device 16x), delivery data to the worker y (worker device 16y), and delivery data to the worker z (worker device 16z). Remember.

  The work result storage unit 38 includes text data that is the result of transcription by worker x (worker device 16 x), text data that is the result of transcription by worker y (worker device 16 y), and worker z (work The text data which is the result of the transcription by the user device 16z) is stored.

  The sentence storage unit 40 stores text data (hereinafter also referred to as “sentence data”) indicating the content of the entire target voice generated by a sentence generation unit 62 described later. The evaluation storage unit 44 stores evaluation results regarding a plurality of workers, which are generated by an evaluation unit 66 described later.

  The control unit 20 includes a request reception unit 50, a conversion unit 52, a division unit 54, an allocation unit 56, a distribution unit 58, a work result reception unit 60, a sentence generation unit 62, a sentence provision unit 64, and an evaluation unit 66. A computer program in which the functions of the plurality of functional blocks are implemented may be stored in the storage unit 22. The processor of the management device 12 may exert the functions of a plurality of functional blocks of the control unit 20 by reading the computer program to the main memory and executing it.

  The request receiving unit 50 receives, from a plurality of user terminals 14, a plurality of request data for requesting transcription of voice. The request receiving unit 50 stores the received request data in the target voice storage unit 30 in association with the user or the user terminal 14 of the request source. The request data received from the user terminal 14a includes target voice data in which a voice related to company A (such as a statement from the president or a voice of a meeting) is recorded. In addition, the request data received from the user terminal 14b includes target voice data in which voice related to company B is recorded. In addition, the request data received from the user terminal 14c includes target voice data in which voice related to company C is recorded.

  The conversion unit 52 converts voice quality (pitch and sound of a plurality of target voice data by converting at least one target voice data of the plurality of target voice data received by the request receiving unit 50 using a known voice conversion function). Pressure, timbre etc.) This makes it difficult for a worker who has heard a plurality of section voice data to distinguish whether or not the original target voices of the section voice data are the same, thereby enhancing the secrecy of the content of the target voice.

  In the embodiment, the converting unit 52 converts the voice quality of the plurality of target voice data received by the request receiving unit 50 so as to be the same as or similar to the voice quality of the dummy voice data. This makes it difficult for a worker who has heard a plurality of section voice data to distinguish whether or not the original target voice of the section voice data is the same, and makes it difficult to identify whether it is a dummy voice or not. It is possible to further enhance the secrecy of the content of the target voice.

  The dividing unit 54 divides the target voice data stored in the target voice storage unit 30 into a plurality of section voice data relating to a plurality of sections. FIG. 3 shows an example of division of audio data. The dividing unit 54 divides the target voice data Aa of company A into three of section voice data Aa-1, section voice data Aa-2, and section voice data Aa-3. Further, the dividing unit 54 divides the target voice data Ab of company A into three of section voice data Ab-1, section voice data Ab-2, and section voice data Ab-3. Similarly, the dividing unit 54 divides the target voice data Ba and the target voice data Bb of company B.

  As described above, in the embodiment, the dummy sound data is divided in advance into a plurality of section sound data. For example, in FIG. 3, the dummy sound data Ca is divided into two, that is, section sound data Ca-1 and section sound data Ca-2. As a modification, the dividing unit 54 may divide the dummy sound data into a plurality of section sound data at the time of dividing the target sound data.

  The dividing unit 54 stores, in the storage unit 22, position information (for example, the order from the beginning, the time position, and the like) in the target sound data before division or in the dummy sound data for each of the plurality of section sound data. For example, the division unit 54 stores, for the section sound data Aa-1, information indicating that it is the first section of the target sound data Aa, and for the section sound data Aa-2, 2 of the target sound data Aa. Information indicating that it is the second section may be stored.

  FIG. 4 shows an example of audio data. The figure shows the waveform of the voice. Specifically, the horizontal axis of the figure shows the elapsed time from the start of the voice, and the vertical axis shows the volume. The dividing unit 54 holds predetermined minimum time and maximum time (in other words, maximum length) which can be taken by the section dividing the voice. The minimum time of the section in the embodiment is 10 seconds (end range start point 70 in FIG. 4), and the maximum time is 20 seconds (end range end point 72 in FIG. 4). The shorter the section, the higher the secrecy of the speech content, but the less accurate the transcription. The minimum time and the maximum time of the section may be determined to be appropriate values by comparing the secrecy of the speech content and the transcription accuracy.

  When the dividing unit 54 determines the end position of one section in the target audio data, the section ends when the volume falls within a range from the predetermined minimum time to the maximum time and the volume is less than the predetermined threshold. Determined as a position. For example, in the example of FIG. 4, a point of time 15.5 seconds after the start of speech is determined as the end position (division point 74) of the section. For the next section, the division unit 54 sets the division point 74 in FIG. 4 as the start position, and within the range of 10 seconds to 20 seconds from the division point 74, the time point at which the volume is less than the predetermined threshold is the next section. Determined as the end position of. Note that the threshold value of the volume may be a volume value considered to be silent, or may be a volume value assumed for a quiet room. For example, the volume threshold may be 0.002 Pascal (40 decibels).

  In the target voice, word breaks and word breaks are likely to decrease in volume. In the embodiment, by setting the position at which the volume is less than the threshold as the end position of the section, it is possible to set a word break or a meaning break as the end position of the section, and to improve the accuracy of transcription.

  Returning to FIG. 2, the assignment unit 56 sets a plurality of sections of section voice data based on the target voice data and section voice data based on the dummy voice data according to the assignment rule stored in the assignment rule storage unit 34. Assign to each of. The assignment unit 56 stores in the delivery data storage unit 36 a set of section voice data based on the target voice data assigned to each worker and section voice data based on the dummy voice data.

  The assignment rules of the embodiment are defined such that a plurality of section voice data to be assigned to one worker are temporally and spatially separated from one another. Specifically, (1) in the case where a plurality of worker devices 16 exist, the assignment rule applies to one worker device 16 among a plurality of section voice data originating from one target voice data. Decide to allocate some interval voice data. That is, the assignment rule prohibits assignment of all section voice data based on one target voice data to one worker device 16. Thereby, the secrecy of the content of the target voice can be enhanced.

  In addition, (2) the assignment rule prohibits assignment of a plurality of section voice data continuous in time in one target voice data to one worker device 16. In other words, the assignment rule prescribes that a plurality of section voice data continuous in time in one target voice data be assigned to different workers. For example, it is defined that the section voice data Aa-1 and the section voice data Aa-2 in FIG. 3 are prohibited from being assigned to the same worker, and are assigned to different workers. This can further enhance the secrecy of the content of the target voice.

  In addition, (3) in the case of providing a plurality of section voice data to one worker device 16, the allocation rule provides a plurality of section voice data originating from one target voice data, It is determined to give priority to providing a plurality of section voice data originating from different target voice data. Since there is a high possibility that the content is not related to the plurality of section voice data originating from different target voice data, the confidentiality of the content of each target voice can be further enhanced.

  Further, (4) in the case where a plurality of section voice data originating from different target voice data is provided to one worker device 16, a plurality of assignment rules originate from different target voice data relating to the same organization. By providing section voice data, it is determined to give priority to providing a plurality of section voice data originating from different target voice data relating to different tissues. Since a plurality of section voice data originating from different target voice data relating to different organizations is more likely to be unrelated to the content, the confidentiality of the content of each target voice can be further enhanced.

  FIG. 5 shows an example of assignment of section voice data. The section voice data shown in the figure corresponds to the section voice data shown in FIG. In the example of the figure, the assignment unit 56 gives the worker x the section voice data Aa-1 originating from the target voice data Aa and the section voice data Ca-1 originating from the dummy voice data Ca. And section voice data Bb-2 originating from the target voice data Bb. In addition, the assignment unit 56 assigns a plurality of section voice data originating from different target voice data of different organizations to the worker y and the worker z, respectively.

  Returning to FIG. 2, the distribution unit 58 provides the section voice data to each worker according to the allocation result by the allocation unit 56. Specifically, distribution unit 58 provides a set of section voice data based on the target voice data stored in delivery data storage section 36 and section voice data based on dummy voice data to worker device 16 of each worker. Do.

  In the embodiment, the distribution unit 58 transmits a web page (hereinafter also referred to as “work page”) for performing the transcription work to the plurality of worker devices 16 and causes the plurality of worker devices 16 to display the web page. The distribution unit 58 provides a work page for the worker x to the worker device 16x, provides a work page for the worker y to the worker device 16y, and a work page for the worker z to the worker device 16z. provide. The distribution unit 58 may notify the worker device 16 of each worker of the URL of the work page for each worker by e-mail or the like.

  The distribution unit 58 assigns section voice data assigned to the worker x by the assignment unit 56 to data of a work page for the worker x (section voice data Aa-1, section voice data Ca-1, and so on in the example of FIG. 5). Section voice data Bb-2) is included. Similarly, the distribution unit 58 includes the section voice data assigned to the worker y (worker z) by the assigning unit 56 in the data of the work page for the worker y (worker z). In addition, the distribution unit 58 arranges a button for reproducing each section voice data, an area for inputting a text as a result of transcribing the sound of each section voice data, and a transmission button on the work page for each worker. .

  The work result receiving unit 60 receives the work result by each worker transmitted from the worker device 16 of each worker. In the embodiment, the work result receiving unit 60 receives the transcription result by the worker x input to the work page for the worker x, and the transcription result by the worker y input to the work page for the worker y Is accepted, and the transcription result by the worker z input to the work page for the worker z is received. The work result receiving unit 60 stores the work result of each worker in the work result storage unit 38.

  FIG. 6 shows an example of the work result by the worker. The figure shows the work result based on the assignment of FIG. For example, the work result by the worker x includes text data Aa-1, text data Ca-1, and text data Bb-2. The text data Aa-1 is obtained by transcribing the voice of the section voice data Aa-1 originating from the target voice data Aa. The text data Ca-1 is obtained by transcribing the voice of the section voice data Ca-1 originating from the dummy voice data Ca. The text data Bb-2 is obtained by transcribing the voice of the section voice data Bb-2 originating from the target voice data Bb.

  Returning to FIG. 2, the text generation unit 62 generates text based on the section voice data originating from the target voice data among the text data received by the work result reception unit 60 and stored in the work result storage unit 38. Using the translated text data, text data is generated by transcribing the entire target voice. When generating sentence data, the sentence generation unit 62 does not use text data generated by transcribing based on interval speech data originating from dummy speech data.

  The sentence generation unit 62 combines text data based on a plurality of section voice data in accordance with position information of each section voice data stored in the storage unit 22 by the division unit 54 (that is, position information in target voice data). Generate sentence data by. The sentence generation unit 62 stores sentence data for a certain target speech in the sentence storage unit 40 in association with the user (or the user terminal 14) that has requested transcription of the target speech.

  The text data Aa-1 included in the work result of the worker x in FIG. 6 is the text of the section voice data Aa-1 corresponding to the first section of the target voice data Aa. The text data Aa-2 included in the work result of the worker y in FIG. 6 is the text of the section voice data Aa-2 corresponding to the second section of the target voice data Aa. The text data Aa-3 included in the work result of the worker z in FIG. 6 is the text of the section voice data Aa-3 corresponding to the third section of the target voice data Aa. The sentence generation unit 62 synthesizes the text data Aa-1, the text data Aa-2, and the text data Aa-3 in this order to generate the sentence data Aa in which the entire target speech data Aa is converted into text.

  Returning to FIG. 2, the sentence providing unit 64 transmits the sentence data stored in the sentence storage unit 40 to the user (user terminal 14) that has requested transcription. For example, the sentence providing unit 64 transmits the sentence data Aa in which the target voice data Aa illustrated in FIG. 6 is transcribed to the user a (user terminal 14a) that has requested the transcription. The sentence providing unit 64 receives, from the plurality of sentence data stored in the sentence storage unit 40, the sentence data associated with the user a when the provision request for the sentence data is received from the user terminal 14a. You may transmit to the user terminal 14a.

  The evaluation unit 66 is the correct answer data stored in advance in the correct answer storage unit 42 and the transcription result of the dummy voice by each worker (text data received by the work result accepting unit 60 and stored in the work result storage unit 38) Evaluate each worker by comparing with. For example, the evaluation unit 66 compares the correct answer data of the section speech Ca-1 originating from the dummy speech data Ca with the text data Ca-1 which is a transcription result of the section speech Ca-1 by the task x. The worker x is evaluated by

  In the embodiment, the evaluation unit 66 extracts morphemes constituting correct data and morphemes constituting text data of the work result by morpheme analysis, and as the number of morphemes coincident with each other increases, the operator converts It is evaluated that the accuracy is high, and the worker is given a high evaluation value. Note that the evaluation unit 66 may refer to the synonym dictionary, and even if the morpheme forming the correct data and the morpheme forming the text data of the work result do not match, it may be regarded as a match if it is a synonym. Thus, in the embodiment, the worker can be objectively evaluated based on the transcription result of the dummy voice.

  The evaluation unit 66 stores the evaluation results (evaluation values) of each of the plurality of workers in the evaluation storage unit 44. The management apparatus 12 may further include an evaluation result output unit (not shown) that provides the external device with the evaluation result of each worker stored in the evaluation storage unit 44. The external device in this case may be, for example, a terminal of a person in charge of making a charge negotiation with the worker or making a contract.

  The operation of the transcription system 10 having the above configuration will be described. A plurality of users of the transcription system 10 upload target voice data from the user terminal 14 to the management apparatus 12 respectively. The request receiving unit 50 of the management device 12 receives a plurality of target voice data transmitted from a plurality of user terminals 14. The conversion unit 52 of the management device 12 converts the voice quality of a plurality of target voice data into a predetermined reference voice quality (in the embodiment, the voice quality is the same as the dummy voice data, and may be synthetic voice quality).

  The dividing unit 54 of the management device 12 divides each of the plurality of target voice data into a plurality of section voice data. The assignment unit 56 of the management device 12 assigns a set of section voice data of the target voice data and section voice data of the dummy voice data to each worker. The distribution unit 58 of the management device 12 presents, to each worker, a set of section voice data of the target voice data and section voice data of the dummy voice data on the web page for each worker.

  The worker reproduces the section voice data assigned to him / her in the web page directed to himself / herself, and inputs text indicating the voice contents in a predetermined area of the web page. When the worker presses the web page transmission button, the worker device 16 transmits the text data input to the predetermined area by the worker to the management device 12.

  The work result receiving unit 60 of the management device 12 has text data obtained by transcribing the section voice of the target voice data transmitted from the worker device 16 of each worker and text data obtained by transcribing the section voice of the dummy voice data. Accept. The sentence generation unit 62 of the management device 12 synthesizes text data obtained by transcribing the section voice of the target voice data transmitted from the worker device 16 of each worker, and a sentence obtained by converting the voice of the entire target voice data into text Generate data.

  The sentence providing unit 64 of the management device 12 transmits the sentence data corresponding to each target voice data to the user terminal 14 that has uploaded each target voice data. The user obtains the sentence data corresponding to the target voice data uploaded by the user and proceeds with the task. The evaluation unit 66 of the management device 12 determines the evaluation value of each worker based on the transcription result of the dummy voice data whose content is previously determined.

  The present invention has been described above based on the embodiments. It is understood by those skilled in the art that this embodiment is an exemplification, and that various modifications can be made to the combination of each component described in the embodiment and each processing process, and such a modification is also within the scope of the present invention. It is about Hereinafter, modified examples are shown.

  A first modification will be described. The dividing unit 54 of the management device 12 divides the target voice data to generate the first section voice data of the first section and the second section voice data of the second section immediately after the first section. , A part of the first section and a part of the second section may be overlapped. In other words, the division unit 54 may provide a time region that is an extension of the first section and the second section.

  FIG. 7 shows an example of division of audio data. Here, the target voice data 80 is divided into four of the section voice data 82a, the section voice data 82b, the section voice data 82c, and the section voice data 82d. The dividing unit 54 provides an overlap period 84 a and an overlap period 84 b in the section voice data 82 a and the section voice data 82 b. Further, the dividing unit 54 provides an overlapping period 84c and an overlapping period 84d in the section audio data 82b and the section audio data 82c. Further, the dividing unit 54 provides an overlapping period 84 e and an overlapping period 84 f in the section audio data 82 c and the section audio data 82 d. Here, each of the overlapping period 84 a to the overlapping period 84 f is 2.5 seconds.

  In the example of FIG. 7, the section voice data 82 a is a voice of a section of 15 seconds from the start point of the target voice data 80. The end position of the section may be determined by the method described in the embodiment. In this section, 5 seconds before the end is the overlapping period (overlapping period 84 a + overlapping period 84 b). The section sound data 82 b is sound of a section from 10 seconds to 25 seconds after the start point of the target sound data 80. In this section, 5 seconds after the start and 5 seconds before the end are overlapping periods. Although the section voice data 82a and the section voice data 82b are assigned to different workers, voices of the overlapping period 84a and the overlapping period 84b are transcribed by both different workers.

  The section voice data 82c is a voice of a section from 20 seconds to 35 seconds after the start point of the target voice data 80. In this section, 5 seconds after the start and 5 seconds before the end are overlapping periods. The section sound data 82 d is sound of a section from 30 seconds to 45 seconds from the start point of the target sound data 80. In this section, 5 seconds after the start is the overlapping period.

  The sentence generation unit 62 generates a predetermined number of overlapping text data of the first interval audio data (for example, interval audio data 82a) and the text data of the second interval audio data (for example, interval audio data 82b) that are continuous in time. The two text data are combined so that the characters of (or a predetermined number of morphemes) may match.

  In addition, the sentence generation unit 62 adopts the text of the section voice data far from the end as the text data in the overlapping period. In other words, the text data (for example, a predetermined number of characters and morphemes) corresponding to the end of each section voice data is not reflected in the text data of the synthetic word. For example, the sentence generation unit 62 adopts text data based on the section audio data 82a for the overlapping period 84a, while adopting text data based on the section audio data 82b for the overlapping period 84b. Similarly, the sentence generation unit 62 adopts text data based on the section audio data 82b for the overlapping period 84c, while adopting text data based on the section audio data 82c for the overlapping period 84d.

  The present inventors have recognized that the accuracy of text conversion decreases at the start and end of interval speech data in carrying out a proof of concept (PoC) for transcription. Therefore, in the present modification, an overlapping period is provided for the first interval audio data and the second interval audio data that are temporally continuous, and the text data of the first interval audio data and the text data of the second interval audio data The accuracy of the text data can be improved by reflecting the portion of the text data that is considered to have a high degree of accuracy.

  A second modification will be described. In the above embodiment, the person hears the target voice and the dummy voice and transcribes them. However, as a modification, the computer (at least some of the worker devices 16) may automatically execute the transcription process. In this case, the distribution unit 58 calls the transcription request API that the worker device 16 discloses on the network, and transmits one or more section voice data (for example, distribution data in FIG. 5) to the worker device 16 May be The work result receiving unit 60 may receive text data of a transcription result as a return value of the transcription request API of the worker device 16.

  A third modified example related to the second modified example will be described. The transcription may be performed by both computer transcription and human transcription. Specifically, the distribution unit 58 of the management device 12 first transmits one or more section voice data (for example, the distribution data of FIG. 5) to the first worker device that automatically executes the transcription process, The result receiving unit 60 may obtain the result of the transcription process from the first worker device. Next, the distribution unit 58 transmits the result of the transcription process by the first worker device to the second worker device that manually performs the transcription, and the work result receiving unit 60 Here, the result of the inspection / editing may be acquired from the second worker device. According to this configuration, a person is in charge of checking and editing the result of the transcription by the computer, so that the accuracy of the transcription can be increased while suppressing the labor cost.

  A fourth modification will be described. Although not mentioned in the above embodiment, the assigning unit 56 may assign the transcription of the section voice data in preference to the worker with the higher evaluation value stored in the evaluation storage unit 44. In addition, the distribution unit 58 may distribute section voice data with priority given to workers with higher evaluation values stored in the evaluation storage unit 44. In other words, section voice data may be allocated or distributed to a worker with a relatively high evaluation value stored in the evaluation storage unit 44 in preference to a worker with a relatively low evaluation value. . This makes it easy to increase the accuracy of the transcription.

  A fifth modification will be described. Although not mentioned in the above embodiment, the evaluation of the worker by the evaluation unit 66 may be performed before the text data generation by the text generation unit 62. When the evaluation value of a certain worker is less than a predetermined threshold value, the sentence generation unit 62 generates sentence data using a transcription result (text data) by the worker (hereinafter referred to as “low evaluator”). You may cancel In this case, the allocating unit 56 may reallocate the section voice data allocated to the low evaluator to another worker (worker whose evaluation value is equal to or higher than the threshold). The distribution unit 58 may provide the section voice data provided to the low rater to the other worker and request the transcription. This makes it possible to avoid providing the transcription result with low accuracy to the user, and to further increase the transcription accuracy.

  A sixth modification will be described. Although not mentioned in the above embodiment, it is desirable that the allocation unit 56 provide different dummy sound data (at least different as section sound data) for at least a predetermined period in the distribution data for the same worker. For example, the assignment unit 56 may save identification information of the assigned dummy sound data (the section sound data thereof) for each worker, and change the dummy sound data (section sound data) each time a request is made. This makes it difficult for the operator to determine the dummy speech, and the confidentiality of the content of the target speech can be enhanced.

  A seventh modification will be described. The management apparatus 12 of the above embodiment provides a web page for each worker to the worker device 16, and causes each worker to reproduce the section voice data on the web page for each worker, and each work Was asked to input the transcript result of the section voice data. In a modification, the distribution unit 58 of the management device 12 may transmit, to the worker device 16, encrypted data obtained by encrypting the section voice data of the target voice and the section voice encrypted data of the dummy voice. The work result receiving unit 60 of the management device 12 may receive, from the worker device 16, encrypted data obtained by encrypting text data of a transcription result by each worker.

  An eighth modification will be described. How many dummy voice interval voice data should be inserted into the target voice interval voice data, or the ratio of the target voice interval voice data to the dummy voice interval voice data to be assigned to the worker is required by the user It may be determined by the confidentiality strength of the target voice. That is, the allocation rule may be set such that the number of section voice data of dummy voice inserted into the distribution data increases as the required level of confidentiality strength increases. Alternatively, the allocation rule may be defined such that the ratio of the section voice data of the dummy voice to the section voice data of the target voice in the distribution data is high. In addition, since the secrecy is enhanced as the above insertion number or ratio is larger, the selling price of the transcription service may be set higher.

  A ninth modification will be described. The configuration of the transcription system 10 described in the above embodiment is merely an example, and it is a matter of course that there is no limitation on the physical configuration (the number of cases, etc.). For example, the function of dividing original voice data and providing section voice data to the worker device 16 and the function of collecting a result of transcription by the worker and generating text data and providing it to the user are different. It may be realized by an apparatus.

  Any combination of the above-described embodiments and variations is also useful as an embodiment of the present invention. The new embodiments resulting from the combination combine the effects of each of the combined embodiments and variations. Further, it is understood by those skilled in the art that the functions to be performed by the respective constituent elements described in the claims can be realized by a single member of each constituent shown in the embodiment and the modification or a combination thereof.

  10 transcription system, 12 management devices, 14 user terminals, 16 worker devices, 52 conversion units, 54 division units, 58 distribution units, 60 work result reception units, 62 sentence generation units, 64 sentence provision units, 66 evaluation units.

Claims (6)

A first storage unit for storing target voice data in which a target voice of transcription is recorded;
A second storage unit for storing dummy voice data in which a dummy voice is recorded;
A division unit that divides the target speech data into a plurality of section speech data relating to a plurality of sections;
A providing unit for providing a set of at least one of the plurality of section voice data and the dummy voice data to an external device;
A receiving unit for receiving text data transcribed based on at least one of the plurality of section voice data and text data transcribed based on the dummy voice data;
A generation unit that generates text data in which the target speech is transcribed using text data transcribed from at least one of the plurality of section voice data among the text data accepted by the acceptance unit; ,
A third storage unit storing text data indicating contents of predetermined voice recorded in the dummy voice data;
Evaluation Department,
Equipped with
When providing a plurality of section voice data originating from different target voice data to one external device, the providing unit provides a plurality of section voice data originating from different target voice data relating to the same tissue. Prioritizing giving priority to providing a plurality of interval speech data originating from different target speech data relating to different organizations,
The evaluation unit evaluates an entity that has performed transcription by comparing the text data stored in the third storage unit with the text data that the transcription unit has received the dummy voice data received by the reception unit. And
When the evaluation unit evaluates an evaluation value for a certain subject by the evaluation unit to be lower than a predetermined threshold value, the providing unit provides the section voice data provided for the certain subject to another subject so as to the information processing system characterized that you request to the other entities in place of transcript in the certain entity.   The division unit is characterized in that, when determining the end position of one section in the target audio data, the division section determines a point at which the volume is less than a predetermined threshold within the maximum length of a predetermined section as the end position. The information processing system according to claim 1, wherein   When a plurality of the external devices exist, the providing unit provides a set of partial speech data and dummy speech data among a plurality of the partial speech data to one external device. The information processing system according to claim 1 or 2. 4. The apparatus according to any one of claims 1 to 3 , further comprising: a conversion unit that homogenizes voice quality of the plurality of target voice data by converting at least one target voice data among the plurality of target voice data. Information processing system as described. The division unit divides the target sound data to generate a first section sound data relating to a first section and a second section sound data relating to a second section immediately after the first section. The information processing system according to any one of claims 1 to 4 , wherein a part of one section and a part of the second section are overlapped. Target voice data in which voice of a target of transcription is recorded, dummy voice data in which dummy voice is recorded, and dummy voice text data indicating contents of predetermined voice recorded in the dummy voice data The information processing system to store
Dividing the target sound data into a plurality of section sound data relating to a plurality of sections;
Providing a set of at least one of the plurality of section voice data and the dummy voice data to an external device;
Accepting both text data transcribed based on at least one of the plurality of section voice data and text data transcribed based on the dummy speech data;
Generating text data obtained by transcribing the target voice using text data generated by transcribing based on at least one of the plurality of section voice data among the received text data;
The execution,
When the step of providing provides a plurality of section voice data originating from different target voice data to one external device, the plurality of section voice data originating from different target voice data relating to the same tissue is provided Prioritizing providing, giving priority to providing a plurality of interval speech data originating from different target speech data relating to different organizations,
The information processing system
Evaluating a subject that has performed transcription by comparing dummy speech text data stored in advance with text data obtained by transcribing the dummy speech data, received in the receiving step;
If the evaluation value for a certain subject is lower than a predetermined threshold value in the evaluating step, transcribing the voice data of the area by providing the voice data of the section provided to the one subject to another subject. And D. requesting the other subject instead of the certain subject .

Images