JP2017111339A - Voice reproduction device, voice reproduction method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a voice reproduction device, voice reproduction method, and program that can perform pinpoint check for a voice associated with a desired portion of a character string that is a result of voice recognition.SOLUTION: A display control part 34 displays a character string that is a result of voice recognition. A voice reproduction part 40, according to designation of one or more characters included in the character string that is a result of voice recognition, reproduces the voice from reproduction positions of the one or more characters associated with a position in the character string that is a result of voice recognition.SELECTED DRAWING: Figure 7

Description

  The present disclosure relates to an audio reproduction device, an audio reproduction method, and a program.

  There is a technique for displaying a voice recognition result, which is an answering machine message, as a character string (see Patent Documents 1 to 3). According to such a technique, the message of the answering machine can be visually confirmed.

JP-A-5-347658 JP-A-11-150603 JP 2003-224700 A

  Even if the message of the answering machine can be confirmed by visual observation, it is desirable that, for example, important parts such as date, time, and location can be reproduced to confirm whether or not there is an error in voice recognition.

  It is also desirable to be able to confirm the meaning of the sound by reproducing the sound even in a portion where the speech recognition is not correctly performed and the meaning of the meaning is unknown from the displayed character string.

  For example, in the above-described scenes and the like, it is convenient if the voice associated with the desired portion of the character string that is the voice recognition result can be confirmed pinpointed. I couldn't do anything.

  In view of the above situation, the present disclosure proposes an audio reproduction device, an audio reproduction method, and a program that can pinpoint the voice associated with a desired portion of a character string that is a speech recognition result.

  In order to solve the above problem, an audio reproduction device according to the present disclosure includes a display control unit that displays a character string that is a speech recognition result, and one or more characters included in the character string, An audio reproduction unit that reproduces the audio from a reproduction position associated with the position of the one or more characters in the character string.

  In one aspect of the present disclosure, the sound reproducing unit may specify the nth partial character string from the beginning when the character string is divided into a plurality of partial character strings each including one or more characters. Corresponds to the playback position belonging to the nth partial time from the beginning or the predetermined time before the playback position when the playback time of the voice is divided into a plurality of partial times by the number of the partial character strings. The sound is reproduced from the reproduction position.

  In this aspect, each of the plurality of partial times may be obtained by dividing the audio playback time at equal time intervals.

  The playback position belonging to the nth partial time may be the start playback position of the nth partial time.

  Moreover, in one aspect of the present disclosure, the display control unit highlights a character associated with a portion where the sound is reproduced.

  In addition, the audio reproduction method according to the present disclosure includes a step of displaying a character string that is a speech recognition result, and the designation of the one or more characters included in the character string according to the designation of the one or more characters. Reproducing the sound from a reproduction position associated with a position in a character string.

  In addition, the program according to the present disclosure may be configured to display a character string that is a speech recognition result, in accordance with designation of one or more characters included in the character string, in the character string of the one or more characters. Causing the computer to execute a procedure of reproducing the sound from a reproduction position associated with the position.

It is a figure showing an example of the whole composition of an answering machine system concerning one embodiment of this indication. It is a figure which shows an example of the flow of the procedure for making the answering machine program which concerns on one Embodiment of this indication usable. It is a flowchart which shows an example of the flow of the process performed by the answering machine processing server which received the transfer telephone. It is a figure which shows an example of a message list screen. It is a figure which shows an example of an audio | voice reproduction | regeneration screen. It is a figure which shows an example of an audio | voice reproduction | regeneration screen. It is a figure which shows an example of an audio | voice reproduction | regeneration screen. It is a figure which shows an example of an audio | voice reproduction | regeneration screen. It is a figure which shows an example of an audio | voice reproduction | regeneration screen. 3 is a functional block diagram illustrating an example of functions implemented in a mobile phone terminal according to an embodiment of the present disclosure. FIG. It is a flowchart which shows an example of the flow of the process performed in the mobile telephone terminal which concerns on one Embodiment of this indication.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

  FIG. 1 is a diagram illustrating an example of the overall configuration of an answering machine system 1 proposed in the present disclosure. As shown in FIG. 1, the answering machine system 1 proposed in the present disclosure includes an answering machine processing server 10, a voice recognition server 12, and a plurality of mobile phone terminals 14. The answering machine processing server 10 and the mobile phone terminal 14 are connected to a telephone communication network 16 and the Internet 18. Therefore, the answering machine processing server 10 and the mobile phone terminal 14 and the mobile phone terminals 14 can communicate with each other via the telephone communication network 16 and the Internet 18. The voice recognition server 12 is connected to the Internet 18. Therefore, the voice recognition server 12 can communicate with the answering machine processing server 10 and the like via the Internet 18.

  The answering machine processing server 10 is a server computer that provides services such as an answering machine service.

  The voice recognition server 12 is a server computer that provides a service that receives, for example, voice and generates text such as a character string that is a voice recognition result for the voice. In the present embodiment, for example, the voice recognition server 12 accepts voice of an answering machine message recorded by the answering machine processing server 10. Then, the voice recognition server 12 generates text data including a character string that is a voice recognition result of the voice by executing voice recognition processing on the voice. Then, the voice recognition server 12 transmits the generated text data to the answering machine processing server 10.

  The mobile phone terminal 14 is a terminal such as a smartphone. As shown in FIG. 1, the mobile phone terminal 14 according to the present embodiment includes, for example, a control unit 14a, a storage unit 14b, a communication unit 14c, a touch panel 14d, and a voice input / output unit 14e.

  The control unit 14a is a program control device such as a microprocessor that operates according to a program installed in the mobile phone terminal 14, for example.

  The storage unit 14b is a storage element such as a ROM or a RAM, for example. The storage unit 14b stores a program executed by the control unit 14a.

  The communication unit 14 c is a communication interface such as a mobile phone communication unit for performing voice communication and data communication via the telephone communication network 16 and a wireless LAN module for performing data communication via the Internet 18.

  The touch panel 14d includes, for example, a touch sensor and a display such as a liquid crystal display or an organic EL display, and displays an image generated by the control unit 14a. Further, the user can perform an operation input to the control unit 14a by operating the touch panel 14d. The control unit 14a executes various processes in response to operation inputs on the touch panel 14d.

  The audio input / output unit 14e includes an audio output device such as a headphone or a speaker, for example, and outputs audio or the like represented by audio data received by the communication unit 14c. The voice input / output unit 14e includes a voice input device such as a microphone. For example, the voice input / output unit 14e transmits received voice to another mobile phone terminal 14 via the communication unit 14c.

  By installing the answering machine program according to the present embodiment, the mobile phone terminal 14 according to the present embodiment can display a voice recognition result as a recorded answering machine message on the touch panel 14d as a character string. It is like that.

  Here, an example of the flow of procedures for making the answering machine program according to the present embodiment available will be described with reference to the flowchart shown in FIG.

  First, the user of the mobile phone terminal 14 obtains the URL and serial number from which the answering machine program according to the present embodiment is obtained at a store of the mobile phone terminal 14 (S101).

  Then, the user accesses the URL obtained in the procedure shown in S101 from the mobile phone terminal 14, downloads the answering machine program according to the present embodiment, and installs the answering machine program in the mobile phone terminal 14 (S102).

  Then, the user inputs the serial number obtained in the procedure shown in S101 via the touch panel 14d (S103). Then, the telephone number to be set as the transfer destination for the non-response transfer is displayed on the touch panel 14d (S104).

  Then, when the user operates the mobile phone terminal 14 and sets the telephone number displayed in the procedure shown in S104 as the transfer destination for the no-response transfer (S105), the user uses the answering machine program according to the present embodiment. It becomes possible.

  For example, it is assumed that a user calls a telephone number of the user's mobile phone terminal 14 that can use the answering machine program according to the present embodiment. Hereinafter, a user who makes a call is referred to as a calling user, and a user to whom the calling user calls is called a receiving user.

  If the incoming user's mobile phone terminal 14 does not answer, the call is forwarded to the answering machine processing server 10 as a call addressed to the telephone number set in the procedure shown in S105. In this embodiment, since the serial number and the transfer destination telephone number are associated with each other on a one-to-one basis, the answering machine processing server 10 determines which telephone number is based on the telephone number set as the transfer destination. It is possible to identify whether the addressed call has been transferred.

  Hereinafter, an example of the flow of processing executed by the answering machine processing server 10 that has accepted a forwarded call will be described with reference to the flowchart shown in FIG.

  When the answering machine processing server 10 receives an incoming call from the mobile phone terminal 14 of the calling user (S201), it sends an automatic response message to the mobile phone terminal 14 of the calling user (S202). This automatic response message is output by voice from the voice input / output unit 14e of the mobile phone terminal 14 of the calling user.

  The answering machine processing server 10 specifies the telephone number of the mobile phone terminal 14 used by the calling user based on the incoming call received in the process shown in S201 (S203).

  Thereafter, when the calling user inputs a voice mail message via the voice input / output unit 14 e of the mobile phone terminal 14, the voice of the message is transmitted to the voice mail processing server 10. The answering machine processing server 10 records the voice of the answering machine message (S204).

  Then, the answering machine processing server 10 transmits the voice data recorded in the process shown in S204 to the voice recognition server 12 (S205). When the voice recognition server 12 receives the voice data, the voice recognition server 12 performs voice recognition processing on the voice. Then, the voice recognition server 12 transmits text data including a character string that is a result of voice recognition of the voice to the answering machine processing server 10. The answering machine processing server 10 receives the text data (S206).

  The answering machine processing server 10 transmits the answering machine data including the text data received in the process shown in S206 and the voice data recorded in the process shown in S204 to the mobile phone terminal 14 of the incoming user (S207). In the present embodiment, the answering machine data includes, for example, an incoming / recording notification, a calling user's telephone number, a recording time of the voice, and a playback time of the voice, in addition to the text data and voice data described above. It is assumed that data indicating each of these is included. Then, the processing shown in this processing example is finished.

  Note that, for example, if the recorded voice is found to be silent as a result of voice recognition, the answering machine processing server 10 uses the voice data of the message recorded in the process shown in S204 in the process shown in S207. You do not have to send it. In this way, the amount of data to be transmitted can be reduced. In this case, a message indicating that the recorded voice is silent may be transmitted. Then, the mobile phone terminal 14 may display the message.

  In addition, the answering machine processing server 10 may transmit voice data from which the silent portion has been removed (trimmed) to the mobile phone terminal 14 of the incoming user in order to reduce the amount of data to be transmitted. .

  The mobile phone terminal 14 of the incoming user who has received the answering machine data transmitted in the process shown in S207 stores the answering machine data. Then, the mobile phone terminal 14 of the incoming user notifies the incoming user that there was an incoming call and recording.

  When the incoming user executes processing for starting the answering machine program according to the present embodiment, a message list screen 20 illustrated in FIG. 4 is displayed on the touch panel 14d.

  On the message list screen 20, text images I1 representing text data included in the received answering machine data are arranged in chronological order in the order of the received time. In addition, on the message list screen 20, an outgoing user name image I2, a photograph image I3, a reproduction icon image I4, a reproduction time image I5, and a recording time image I6 are arranged in association with the text image I1.

  The calling user name image I2 is an image representing the name of the calling user, for example. The photograph image I3 is, for example, a photograph image of the calling user. In the present embodiment, for example, in the contact information application installed in the mobile phone terminal 14 of the incoming user, the name and photo managed in association with the telephone number of the calling user indicated in the received answering machine data Is identified. In the present embodiment, an image representing the specified name is arranged on the message list screen 20 as the calling user name image I2, and an image of the specified photo is arranged on the message list screen 20 as the photo image I3.

  The reproduction icon image I4 is an icon image for instructing audio reproduction. The reproduction time image I5 is an image representing the reproduction time indicated in the received answering machine data. The recording time image I6 is an image representing the recording time indicated in the received answering machine data.

  Here, when the incoming user performs an operation of selecting the reproduction icon image I4, such as a tap operation on the reproduction icon image I4, for example, an audio reproduction screen 22 illustrated in FIG. 5A is displayed on the touch panel 14d.

  On the audio reproduction screen 22 shown in FIG. 5A, a text image I1 associated with the selected reproduction icon image I4 is arranged. Further, in the present embodiment, when the voice reproduction screen 22 is displayed, the voice reproduction of the answering machine message associated with the text image I1 arranged on the voice reproduction screen 22 is started. .

  On the audio playback screen 22, a playback time image I5 indicating the playback time of the sound and a playback position image I7 indicating the playback position are arranged. In addition, a seek bar image I8 and various operation images I9 are arranged on the audio reproduction screen 22. The called user can change the sound reproduction position by operating the seek bar image I8. The incoming user can operate the operation image I9 to perform operations such as fast forward, rewind, stop, playback, and double speed of the voice.

  In the present embodiment, during the reproduction of the voice of the message, the character associated with the reproduction position is highlighted in the character string represented by the text image I1. In FIG. 5A, the highlighted character is expressed as a character surrounded by the cursor C.

  Note that characters associated with the syllable or phoneme being played back need not be highlighted. For example, the highlighted character may be changed every time the reproduction time is divided by the number of characters included in the character string represented by the text image I1. Specifically, for example, when the reproduction time is T1 seconds and the number of characters included in the character string represented by the text image I1 is N1, the n1st character from the top is ((n1-1) × T1 / N1). ) It may be highlighted when the playback position is from the second to (n1 × T1 / N1) seconds.

  Here, as shown in FIG. 5B, for example, it is assumed that the receiving user performs an operation of designating a character (for example, “8” which is the 60th character from the top) indicating the time when the calling user visits the Toranomon branch office. Then, as shown in FIG. 5C, the reproduction position is changed according to the position of the designated character in the character string represented by the text image I1. Then, the voice of the message is reproduced from the reproduction position after the change.

  For example, when the reproduction time is T1 seconds and the number of characters included in the character string represented by the text image I1 is N1, it is assumed that the n1st character from the beginning is designated. In this case, in this embodiment, for example, the voice of the message is reproduced from the reproduction position of (((n1-1) × T1 / N1) −Δ) seconds. Note that Δ is a predetermined offset value, for example, 2 seconds here. In the example of FIG. 5C, since N1 = 132, T1 = 24 seconds, and n1 = 60, the sound is reproduced from the reproduction position of (((60-1) × 24/132) -2) = 8.72 seconds. The Rukoto. As described above, in the present embodiment, the voice of the message is reproduced from a predetermined time before the reproduction position associated with the designated character.

  The value of Δ may be 0. In this case, the voice of the message is reproduced from the reproduction position associated with the designated character. For example, when N1 = 132, T1 = 24 seconds, and n1 = 60, playback is started from the playback position of ((60-1) × 24/132) = 10.72 seconds. Further, a portion of a predetermined time (for example, 5 seconds) may be repeatedly reproduced from the reproduction position where the voice of the message is reproduced according to the designation of the character.

  FIG. 6A is a diagram illustrating an example of the audio reproduction screen 22 on which another example of the text image I1 is arranged. When the accuracy of speech recognition is poor, as shown in FIG. 6A, there may be a portion whose meaning is unknown from the character string represented by the text image I1. It is assumed that the incoming user has performed an operation of designating a portion whose content is unknown (in FIG. 6A, for example, “company” which is the 59th character from the top). Then, also in this case, as shown in FIG. 6B, the reproduction position is changed according to the position of the designated character in the character string represented by the text image I1, and the voice of the message is reproduced from the reproduction position after the change. The In the example of FIG. 6B, since N1 = 117, T1 = 24 seconds, and n1 = 59, the sound is reproduced from the reproduction position of (((59-1) × 24/117) -2) = 9.89 seconds. The Rukoto. It should be noted that a portion of the character string represented by the text image I1 that is poor in accuracy of speech recognition may be highlighted. For example, a portion with poor voice recognition accuracy may be displayed in a color different from that of other characters.

  According to the present embodiment, the receiving user can know the content of the answering machine message recorded by the calling user by viewing the text image I1. In addition, in the present embodiment, by specifying a character included in the character string represented by the text image I1, sound is reproduced from a reproduction position associated with the character.

  For example, if the playback time is T1 seconds and the number of characters included in the character string represented by the text image I1 is N1, and the n1st character from the beginning is specified, the syllable or phoneme representing the character is (( There is a high possibility of playback at a playback position of (n1-1) × T1 / N1) seconds. Based on this, in the present embodiment, when the nth character from the head is specified, the syllable or phoneme can be heard with a margin (((n1-1) × T1 / N1) −Δ). Audio is played from the second playback position. As described above, the sound may be reproduced from the reproduction position of ((n1-1) × T1 / N1) seconds where the syllable or phoneme representing the character is highly likely to be reproduced. In this way, in the present embodiment, the user wants to check the character string represented by the displayed text image I1, such as an important part such as date, time, place, or a part whose meaning is unknown. The voice of the part can be confirmed pinpoint.

  In this embodiment, since the playback position can be specified by the simple method as described above, the playback position of the syllable or phoneme constituting the voice of the message is associated with the character represented by the syllable or phoneme. There is no need to do. For this reason, it is not necessary for the answering machine processing server 10 to transmit to the mobile phone terminal 14 data indicating the correspondence between the reproduction positions of the syllables and phonemes constituting the voice of the message and the characters represented by the syllables and phonemes. Therefore, the communication amount corresponding to the data is saved.

  Hereinafter, the function of the mobile phone terminal 14 according to the present embodiment and the processing executed by the mobile phone terminal 14 according to the present embodiment will be further described with a focus on reproduction of voice at a pinpoint according to the designation of characters. Note that the mobile phone terminal 14 according to the present embodiment plays a role as an audio reproduction device that reproduces audio associated with a designated character.

  FIG. 7 is a functional block diagram showing an example of functions implemented in the mobile phone terminal 14 according to the present embodiment. Note that not all of the functions shown in FIG. 7 need be implemented in the mobile phone terminal 14 according to the present embodiment, and functions other than the functions shown in FIG. 7 may be implemented.

  As shown in FIG. 7, the cellular phone terminal 14 according to the present embodiment functionally includes, for example, an answering machine data receiving unit 30, an answering machine data storage unit 32, a display control unit 34, a designation receiving unit 36, a reproduction position. A determination unit 38 and an audio reproduction unit 40 are included. The answering machine data receiving unit 30 is mainly implemented by the communication unit 14c. The answering machine data storage unit 32 is mainly implemented by the storage unit 14b. The display control unit 34 and the designation receiving unit 36 are mainly implemented by the control unit 14a and the touch panel 14d. The reproduction position determination unit 38 is mainly implemented by the control unit 14a. The audio reproducing unit 40 is mainly implemented with a control unit 14a and an audio input / output unit 14e.

  The above functions are implemented by executing, in the control unit 14a, a program (an answering machine program according to the above-described embodiment) that is installed in the mobile phone terminal 14 that is a computer and that includes instructions corresponding to the above functions. The This program is supplied to the mobile phone terminal 14 via a computer-readable information storage medium such as an optical disc, a magnetic disc, a magnetic tape, a magneto-optical disc, or a flash memory, or via the Internet.

  In this embodiment, for example, the answering machine data receiving unit 30 receives the answering machine data transmitted by the answering machine processing server 10 in the process shown in S207 of FIG.

  In the present embodiment, the answering machine data storage unit 32 stores, for example, the answering machine data received by the answering machine data receiving unit 30.

  In the present embodiment, for example, the display control unit 34 displays the message list screen 20 shown in FIG. 4 and the voices shown in FIGS. 5A to 5C, FIG. 6A, and FIG. The reproduction screen 22 is displayed on the touch panel 14d. Further, as described above, the display control unit 34 may highlight the character associated with the portion where the sound is reproduced.

  In the present embodiment, for example, the designation receiving unit 36 receives designation of one or more characters included in a character string that is a speech recognition result. In the present embodiment, for example, when a tap operation is performed on the touch panel 14d, the designation receiving unit 36 is arranged at the position where the character arranged at the tapped position or the distance from the tapped position is the shortest. Is accepted as the specified character.

  Note that the designation receiving unit 36 may receive the designation of any one of a plurality of partial character strings obtained by dividing the character string represented by the text image I1 represented by the text image I1. Here, the partial character string refers to a character string represented by the text image I1 divided by a predetermined unit such as a sentence unit, a word unit, a line unit, or a predetermined number of character units. In the present embodiment, each partial character string is composed of one or more characters. For example, when the character string represented by the text image I1 is divided into a plurality of partial character strings in units of one character, each of the plurality of partial character strings includes one character. The number of characters included in each of the plurality of partial character strings may be the same or different.

  In the present embodiment, for example, the reproduction position determination unit 38 associates the position of the one or more characters with the position in the character string in accordance with the designation of one or more characters included in the character string that is the speech recognition result. The playback position to be played is determined as a playback position for playing back the sound.

  Here, for example, it is assumed that the designation receiving unit 36 receives the designation of the nth partial character string from the top. In this case, the playback position determination unit 38 determines the playback position belonging to the nth partial time from the beginning when the playback time is divided into a plurality of partial times by the number of partial character strings as the playback position for playing back sound. May be. Alternatively, in this case, the playback position determination unit 38 determines a playback position corresponding to a predetermined time before the playback position belonging to the nth partial time from the beginning (for example, 2 seconds before) as a playback position for playing back sound. Also good. Here, the reproduction position belonging to the nth partial time from the beginning may be, for example, the reproduction position at the beginning of the nth partial time from the beginning. Each of the plurality of partial times may be obtained by dividing the reproduction time at equal time intervals.

  For example, when the playback time is T2 seconds and the number of partial character strings is N2, it is assumed that the n2th partial character string from the beginning is designated. In this case, for example, a reproduction position of (((n2-1) × T2 / N2) −Δ) seconds may be determined as a reproduction position for reproducing sound. For example, as shown in FIG. 6A, the character string represented by the text image I1 is 8 lines, and the partial character string is obtained by dividing the character string in line units. In this case, T2 = 24 seconds and N2 = 8. Here, for example, when a partial character string on the third line is designated, a reproduction position of (((3-1) × 24/8) -2) = 4 seconds is determined as a generation position for reproducing sound. Also good.

  In the present embodiment, for example, in accordance with the designation of one or more characters included in the character string that is the speech recognition result, the voice reproducing unit 40 is associated with the position of the one or more characters in the character string. The sound is reproduced from the reproduction position. In the present embodiment, for example, the audio reproduction unit 40 reproduces audio from a reproduction position determined as a reproduction position by which the reproduction position determination unit 38 reproduces audio. In addition, the voice reproduction unit 40 may repeatedly reproduce a portion of a predetermined time (for example, 5 seconds) from the reproduction position where the voice of the message is reproduced according to the designation of the character. In addition, the audio reproduction unit 40 may repeatedly reproduce from the reproduction position associated with the start point to the reproduction position associated with the end point according to the designation of the start point and end point characters.

  In the present embodiment, the display control unit 34 updates the display content of the audio reproduction screen 22 in accordance with the reproduction position that is determined as the reproduction position at which the reproduction position determination unit 38 reproduces the audio.

  Hereinafter, an example of the flow of processing performed in the mobile phone terminal 14 according to the present embodiment when the designation receiving unit 36 receives designation of a partial character string will be described with reference to the flowchart shown in FIG.

  When the designation receiving unit 36 receives designation of a partial character string, first, the reproduction position determining unit 38 identifies the partial character string (S301). Then, the reproduction position determination unit 38 determines a reproduction position for reproducing the sound as described above based on the partial character string specified in the process shown in S301 (S302). Then, the display control unit 34 updates the display content based on the reproduction position determined in the process shown in S302, and the audio reproduction unit 40 reproduces sound from the reproduction position determined in the process shown in S302 (S303). ), The processing shown in this processing example is terminated.

  In addition, this invention is not limited to the above-mentioned embodiment.

  For example, the answering machine data may include data indicating the correspondence between the reproduction positions of the syllables and phonemes constituting the voice of the message and the characters represented by the syllables and phonemes. Then, the playback position determination unit 38 may play back the sound from the playback position of the syllable or phoneme associated with the designated partial character string or a predetermined time before the playback position.

  For example, the application range of the present invention is not limited to the mobile phone terminal 14. The present invention may be applied to a general computer such as a personal computer.

  The specific character strings and numerical values described above and the specific character strings and numerical values in the drawings are examples, and are not limited to these character strings and numerical values.

  DESCRIPTION OF SYMBOLS 1 Answering machine system, 10 Answering machine processing server, 12 Voice recognition server, 14 Mobile phone terminal, 14a Control part, 14b Storage part, 14c Communication part, 14d Touch panel, 14e Voice input / output part, 16 Telephone communication network, 18 Internet, 20 message list screen, 22 voice playback screen, 30 answering machine data receiving unit, 32 answering machine data storage unit, 34 display control unit, 36 designation receiving unit, 38 playback position determining unit, 40 voice playback unit.

Claims (7)

A display control unit for displaying a character string that is a speech recognition result;
An audio reproduction unit that reproduces the audio from a reproduction position associated with a position of the one or more characters in the character string in accordance with designation of one or more characters included in the character string;
A sound reproducing device comprising: When the nth partial character string from the top in the case where the character string is divided into a plurality of partial character strings each composed of one or a plurality of characters, the sound reproduction unit reproduces the sound When the time is divided into a plurality of partial times by the number of partial character strings, the sound is reproduced from a reproduction position belonging to the nth partial time from the beginning or a reproduction position corresponding to a predetermined time before the reproduction position. Let
The sound reproducing apparatus according to claim 1, wherein Each of the plurality of partial times is obtained by dividing the reproduction time of the sound at equal time intervals.
The sound reproducing apparatus according to claim 2, wherein The playback position belonging to the nth partial time is the first playback position of the nth partial time.
The audio reproducing apparatus according to claim 2 or 3, wherein The display control unit highlights a character associated with a portion where the sound is reproduced,
The sound reproducing device according to any one of claims 1 to 4, wherein Displaying a character string that is a speech recognition result;
Reproducing the sound from a reproduction position associated with a position of the one or more characters in the character string in accordance with designation of one or more characters included in the character string;
An audio reproduction method comprising: The procedure to display the character string that is the speech recognition result,
A procedure for reproducing the sound from a reproduction position associated with a position of the one or more characters in the character string in accordance with designation of one or more characters included in the character string;
A program that causes a computer to execute.

Images