JP5426706B2 - Audio recording selection device, audio recording selection method, and audio recording selection program - Google Patents

Description

  Embodiments described herein relate generally to an audio recording selection device, an audio recording selection method, and an audio recording selection program.

  Conventionally, it is known to record business information by voice. The user who recorded the sound can check the business information and the like retroactively by reproducing the sound.

JP 2010-191802 A Japanese Patent No. 4693494

Ling Bao and Stephen S. Intille, "Activity Recognition from User-Annotated Acceleration Data" (USA), Massachusetts Institute of Technology

  However, when the information that is the basis of the business record is recorded with the voice during work instead of the memo and the business record is created from the recorded voice, it is equivalent to the work time (recording time) where the voice was recorded. It was necessary to play the voice over the time.

The voice recording selection apparatus according to the embodiment includes a voice receiving unit, a voice recognition unit, a work recognition unit, a storage unit, an importance level determination unit, and an output unit. The voice receiving unit receives voice from the user . The voice recognition unit recognizes the voice and converts it into a character string. The work recognition unit recognizes work information indicating a specific work for which a start time and an end time are determined. The storage unit stores a correspondence dictionary in which at least one of a character string and an attribute of a character string important in a specific work and an importance level of the character string and the attribute in the specific work are associated with the specific work. The importance level determination unit determines the importance level of the character string converted from the speech received during the period from the disclosure time of the specific task to the end time based on the task information and the correspondence dictionary. The output unit distinguishes and outputs the character string converted from the speech accepted in the period from the disclosure time to the end time according to the importance.

The hardware block diagram of the audio | voice recording selection apparatus concerning embodiment. The functional block diagram of the audio | voice recording selection apparatus concerning embodiment. The graph which shows the process example which an audio | voice recording part performs. The chart which illustrates the division of a nurse's action. The chart which shows the example of a structure of a corresponding model. The figure which shows the structural example of a corresponding dictionary. The 1st example of a display when a display part displays the result determined by the importance determination part. The 2nd display example when a display part displays the result determined by the importance determination part. The flowchart of operation | movement of the audio | voice recording selection apparatus concerning embodiment.

  Embodiments of an audio recording selection apparatus will be described in detail below with reference to the accompanying drawings. FIG. 1 is a hardware configuration diagram illustrating the configuration of an audio recording selection device 1 according to the embodiment. As shown in FIG. 1, the audio recording / selection device 1 has a computer function including, for example, an operation unit 20, a microphone 21, a detection unit 22, a speaker 23, a communication unit 24, a storage unit 25, and a control unit 26. It is a terminal device. For example, the voice recording selection device 1 may be a smartphone or the like.

  The voice recording selection device 1 is worn on the waist of a user such as a nurse engaged in nursing work, records the voice of the user's action along with the time, information that changes according to the user's action, etc. Is detected and stored together with the time. Moreover, the voice recording selection apparatus 1 recognizes (estimates) an action such as a work performed by the user using a detection result of information that changes according to the action of the user. The voice record selection apparatus 1 according to the embodiment will be described by taking an example in which a nurse who performs nursing work is a user. However, the present invention is not limited to this. For example, a work over a certain amount of time of about one day is performed. After completing the above, a person who performs another job for creating a daily job record may be a user.

  Next, each part which comprises the audio | voice recording selection apparatus 1 is explained in full detail. The operation unit 20 includes an input unit 200 that accepts a user's operation input, and a display unit 202 that displays an operation result of the voice recording selection device 1 and the like. In the operation unit 20, the input unit 200 and the display unit 202 may be integrated by a touch panel or the like.

  For example, when the function (or power supply) of the voice recording selection device 1 is turned on, the microphone 21 receives voice and the like. Note that the microphone 21 also has a function of converting received sound into digital data. In addition, the microphone 21 may be configured with a highly directional microphone, and is provided outside the audio recording selection device 1 to reduce the influence of environmental sounds, and is arranged at the user's mouth or the like. It may be configured.

  The detection unit 22 includes, for example, an acceleration sensor 220, a positioning sensor 222, and the like, and detects information that changes according to the user's behavior. The acceleration sensor 220 may be a sensor that realizes the function of the acceleration sensor by, for example, MEMS (Micro Electro Mechanical System). The acceleration sensor 220 detects the movement (acceleration) of the target person by detecting the acceleration that changes according to the movement of the user. For example, the acceleration sensor 220 continuously detects acceleration when the function (or power supply) of the voice recording selection device 1 is turned on.

  The positioning sensor 222 detects the position of the user using, for example, a GPS (Global Positioning System) function. However, the function of the positioning sensor 222 may be replaced by using beacon information including device identification information transmitted from a Wi-Fi (registered trademark) access point or a Bluetooth (registered trademark) device described later. Further, the function of the positioning sensor 222 may be configured using an IMES (Indoor MEssaging System) that enables positioning by issuing a signal corresponding to GPS positioning information from a device installed indoors. Further, the function of the positioning sensor 222 may be realized by using the position information of the base station using a mobile phone communication network.

  Moreover, the detection part 22 may be provided with the pulse detection part which detects the pulse of the user which changes according to a user's action. Furthermore, the detection part 22 may be provided with the temperature detection part which detects the environmental temperature which affects a user's action. That is, the detection unit 22 detects more information that changes according to the user's behavior and other information that affects the user's behavior, and the accuracy of recognition of the user's behavior (work, etc.). May be configured to be improved. The detection unit 22 may be configured such that all or part of the functions are provided outside the audio recording selection device 1 and the detection result is transmitted to the control unit 26.

  The speaker 23 outputs a sound or the like in accordance with a user operation input via the input unit 200. It is assumed that the speaker 23 also has a data conversion function necessary for outputting sound.

  The communication unit 24 is a general-purpose interface that performs communication with the outside, and is configured to be connectable to, for example, wired communication, long-distance wireless communication, and proximity wireless communication. Specifically, the communication unit 24 is connectable to a wireless local area network (LAN) having a plurality of access points whose mutual connectivity is guaranteed by Wi-Fi (registered trademark). The communication unit 24 is configured to perform communication using a mobile phone network, Bluetooth (registered trademark), RFID (Radio Frequency IDentification), USB (Universal Serial Bus), a wired LAN, and the like.

  The storage unit 25 includes a ROM (Read Only Memory), a RAM (Random Access Memory), a flash memory (Flash memory), and the like (not shown). And the memory | storage part 25 memorize | stores the data (corresponding model 34 and corresponding | compatible dictionary 36 mentioned later) etc. which are used when the audio | voice recording selection apparatus 1 executes a program, and the audio | voice recording selection apparatus 1 execute a program. The storage unit 25 also includes information (time data) indicating the time, such as voice received by the microphone 21, acceleration data detected by the acceleration sensor 220, position information (or access point information) detected by the positioning sensor 222, and the like. Remember with. Further, the audio recording selection device 1 is provided with a storage medium 250 such as a memory card that can exchange programs and data with the storage unit 25 in a detachable manner.

  The control unit 26 includes, for example, a CPU 260 and controls each unit constituting the audio recording selection device 1.

  Next, functions of the audio recording selection device 1 will be described. FIG. 2 is a functional block diagram illustrating an overview of the functions of the audio recording selection device 1 according to the embodiment. As shown in FIG. 2, the voice recording selection apparatus 1 includes a voice receiving unit 30, a voice recording unit 31, a voice recognition unit 32, an information detection unit 33, a correspondence model 34, a work recognition unit 35, a correspondence dictionary 36, and an importance determination. Part 37 and output part 38.

  The voice receiving unit 30 includes a microphone 21 (FIG. 1) and the like. When the function (or power supply) of the voice recording selection device 1 is turned on, the voice receiving unit 30 receives the voice and sends it to the voice recording unit 31 and the voice recognition unit 32. Output.

  The voice recording unit 31 includes the control unit 26 and the storage unit 25, and records (stores) the voice received from the voice receiving unit 30 together with the time. FIG. 3 is a graph showing an example of processing performed by the audio recording unit 31. As shown in FIG. 3, the audio recording unit 31 receives audio data of PCM (Pulse Code Modulation) data expressed by, for example, a 16-bit integer value from the audio receiving unit 30. The voice recording unit 31 stores all of the voice data received from the voice receiving unit 30 or only useful voice data.

  In the present embodiment, as shown in FIG. 3, the sound recording unit 31 provides a threshold for the size of received sound data (speech waveform), and if the size of the sound data is less than the threshold, the environmental sound For example, it is considered that the user does not utter a voice because it is not useful data or silence. That is, the voice recording unit 31 sets only voice data whose size is equal to or greater than the threshold as user voice data.

  As described above, the audio recording unit 31 distinguishes the audio data received from the audio receiving unit 30 into the silent period and the useful continuous audio period, and records the audio data divided for each useful continuous audio period together with the time data. To do. That is, the audio data recorded by the audio recording unit 31 includes the start time and end time of the continuous audio period. The method for detecting the voice period is not limited, and the voice data of the user may be detected using short-time power.

  The voice recognition unit 32 includes a control unit 26, a storage unit 25, and the like. The voice recognition unit 32 recognizes voice data received from the voice reception unit 30, converts the voice data into text (character string), and indicates information indicating the time when the voice is emitted. The associated character string is output to the importance determination unit 37. The speech recognition unit 32 includes, for example, an LVCSR (Large-Vocabulary Continuous Speech Recognition) engine. The voice recognition unit 32 may have a word dictionary, an acoustic model, and a language model specialized for the user's work.

  The information detection unit 33 includes, for example, an acceleration detection unit 330 including the acceleration sensor 220 and a position detection unit 332 including the positioning sensor 222, detects the user's acceleration and position, and outputs them to the work recognition unit 35. To do. The information detection unit 33 also detects other information such as a pulse detection unit 334 that detects the pulse of the user and a temperature detection unit 336 that detects the temperature of the environment in which the user is located, and sends the detected information to the work recognition unit 35. It may be configured to output in response to this. The user's pulse can be changed when the user rushes to work, and thus can be used when recognizing the user's work. Further, when the user is a nurse, the temperature of the environment in which the nurse is located can be used when recognizing the work of the nurse that occurs in response to a change in the health condition of the patient.

  The correspondence model 34 is a model used by the work recognition unit 35 to recognize a specific work performed by the user using the detection result of the information detection unit 33, and is stored in the storage unit 25, for example. Hereinafter, a case where the work recognition unit 35 recognizes a specific work performed by the user using the detection results of the acceleration detection unit 330 and the position detection unit 332 will be described as an example.

  First, user behavior categories will be described. FIG. 4 is a chart illustrating an example of nurse behavior classification. As shown in FIG. 4, the nurse's actions are classified into, for example, work, work, and action. It is assumed that the nurse's behavior includes a plurality of tasks such as a patient round-up task and a nursing task. Further, the work refers to a small work that occurs during work. In addition, the movement indicates a specific movement of the user such as “posture change”, “moving hand”, and “stopping hand” that occurs during the work. In other words, as illustrated in FIG. 4, the user's behavior is composed of a plurality of operations (“change of body orientation”, “peeling clothes”, etc.) for work (“patient round-up”), and a plurality of operations. It is assumed that it is represented by a hierarchical structure composed of the following operations (“posture change”, “move the hand”, “solve”, etc.). Work is recognized from a plurality of operations, and work is recognized from a plurality of operations.

  In nursing work, when a patient is examined, the nurse sometimes raises the patient's body using both hands. In this case, it is difficult for the nurse to write on the paper the information necessary for creating a work record of the nursing work. Specifically, when using both hands, it is difficult for a nurse to use a writing instrument to write information such as the size and condition of a patient's wound on paper. Therefore, the nurse records voice in the voice recording selection device 1 instead of writing the information on paper.

  FIG. 5 is a chart showing a configuration example of the correspondence model 34. As shown in FIG. 5, in the correspondence model 34, for example, information on the movement pattern 50, position 52, and time 54 is associated with information on the work name 56. For example, in the correspondence model 34, operations (operation patterns) in the order of “bend over”, “stop”, and “wake up” are performed, and this operation pattern is performed in the “patient room” and the operation pattern is performed. When the time required for the patient is within “1 minute”, it is a model that associates that the work of the nurse as the user was “raise the patient's body”. Further, the correspondence model 34 is configured such that a specific work is associated with the repetition of the operation, the occurrence ratio of the operation within a certain time, and the like. Note that when the user's action is associated with a specific work by the correspondence model 34, any one of the work names indicated in the work name 56 is determined.

  The work recognition unit 35 (FIG. 2) is configured by the control unit 26 and the like, accepts acceleration data associated with the time detected by the acceleration detection unit 330, and first recognizes the operation associated with the user's time. To do. Next, the work recognition unit 35 associates the user's time using the correspondence model 34 with the position data in which the recognized motion is associated with the user's time detected by the position detection unit 332. Recognize work. That is, the work recognition unit 35 also recognizes the start time and the end time of each operation and work.

  For example, when recognizing the work of “raising the patient's body” shown in FIG. 5, the work recognition unit 35 first recognizes the actions of “bend”, “stop”, and “wake up”. Here, when the start time of the acceleration data corresponding to the “creasing” operation is time ts, the work recognition unit 35 sets the start time of the “creasing” operation as time ts.

  The work recognizing unit 35 recognizes the “stop” and “wake up” operations following the “bend” operation, and when the end time of the acceleration data corresponding to the “wake up” operation is the time te, Is the time te.

  Then, when the work recognition unit 35 recognizes the series of operations of “bend”, “stop”, and “wake up” as the work of “raise the patient's body”, the work recognition unit 35 sets the start time of the work of “raise the patient's body” Time ts is assumed, and work end time is assumed to be time te.

  The correspondence dictionary 36 (FIG. 2) is used by the importance level determination unit 37 to determine the importance level of the voice data recognized by the voice recognition unit 32 in the specific work recognized by the work recognition unit 35. A dictionary, for example, stored in the storage unit 25.

  FIG. 6 is a chart showing a configuration example of the correspondence dictionary 36. In the correspondence dictionary 36, at least one of a character string (for example, a keyword) and a character string attribute important in a specific work is associated with the specific work. In FIG. 6, a character string enclosed by “<” and “>”, such as “<number>”, is an arbitrary character string as long as it is a keyword (a keyword having the same attribute) belonging to the class indicated by the character string. Represents that they are associated with each other.

  For example, an arbitrary character string indicating a number is associated with the portion indicated by “<number>”. In other words, the correspondence dictionary 36 generates a voice of “10 cm” when the recognized character string “10 cm” is associated with the period recognized by the work “measure the size of the wound”. The character string “10 centimeters” is determined to be a character string having high importance. In addition, since the importance level of the voice (character string) varies depending on the type of work, the correspondence dictionary 36 enables determination that the importance level varies depending on the work associated with the same character string.

  Note that the character strings listed as keywords in FIG. 6 are enumerated together with examples of use or modifications, or defined for use or modification so that they can be associated with work even if they are used or modified. Also good. In FIG. 6, the correspondence dictionary 36 associates at least one of a character string and a character string attribute having the same level of importance with respect to a specific work. At least one of a plurality of levels of character strings and character string attributes having different levels of importance may be distinguished and associated.

  The importance level determination unit 37 (FIG. 2) includes the control unit 26 and the like. The importance level determination unit 37 uses the correspondence dictionary 36 to determine the importance level as a character string of the voice data recognized by the voice recognition unit 32 in the specific work (work period) recognized by the work recognition unit 35. The determination result is output to the output unit 38. For example, when the speech recognition unit 32 recognizes the character string “10 centimeters” for the period in which the work recognition unit 35 recognizes the work “measure the size of the wound”, the importance level determination unit 37 corresponds to the correspondence dictionary. According to 36, the character string “10 centimeters” is determined to have high importance.

  On the other hand, even if the voice recognition unit 32 recognizes the character string “10 centimeters”, the importance determination unit 37 does not recognize that the work recognized by the work recognition unit 35 is a work “measures the size of a wound”. It is not determined that the importance of the character string “10 cm” is high.

  The output unit 38 receives and outputs the result determined by the importance level determination unit 37. Here, the output unit 38 may be the display unit 202 shown in FIG. That is, the output unit 38 may output the result determined by the importance level determination unit 37 by displaying the result.

  The output unit 38 may be the communication unit 24 (FIG. 1). That is, the output unit 38 may output the result determined by the importance level determination unit 37 by transmitting the result to, for example, an external PC (Personal Computer). Then, the user may display or process the determination result of the importance determination unit 37 using an external PC.

(First display example)
FIG. 7 is a first display example when the display unit 202 displays the result determined by the importance level determination unit 37. As illustrated in FIG. 7, when the display unit 202 receives the determination result of the importance determination unit 37, the display unit 202 displays the recognized (estimated) work in time series. The display unit 202 arranges the character strings corresponding to the tasks displayed in time series so that the tasks correspond to the character strings, and displays them by arranging them from the top to the bottom in the order in which, for example, sounds are emitted.

  Further, the display unit 202 distinguishes and displays character strings according to importance for each specific work. For example, the display unit 202 displays the character string 102 (“red”) and the character string 104 (“10 centimeters”) with high importance in “Room 304_Measure the size of the wound” (operation A). The character string 100 and the character string 106 are highlighted and displayed. The display unit 202 emphasizes the display by changing the color of the character string or changing the thickness or size. In addition, the display unit 202 displays a highly important character string 110 (“◯◯ (medicine name)”) and character string 114 (“gauze pasted”) in “Room 304_Applying medicine” (operation B). Is highlighted with respect to the character string 108, the character string 112, and the character string 116.

  Here, both the character string 102 in the work A and the character string 108 in the work B are character strings having a color attribute of <color> and indicating red. However, even if the importance of the character string indicating red in the work A is high, the importance of the character string indicating red in the work B is not high, so the display unit 202 displays the character string 102 and the character string 108. The display is different.

  Note that the voice recording selection device 1 (FIG. 2) has a function of outputting an operation reception unit 40 having a function of accepting a user's operation input and a function of outputting a voice, as exemplified by the hardware configuration in FIG. The audio output unit 41 may be included.

  In addition, the display unit 202 displays only a character string corresponding to one task selected from the plurality of tasks via the operation receiving unit 40, and responds to an operation input via the operation receiving unit 40. Any selected display may be configured.

(Second display example)
FIG. 8 is a second display example when the display unit 202 displays the result determined by the importance level determination unit 37. Note that the second display example shown in FIG. 8 may be displayed on an external PC or the like via the communication unit 24 (FIG. 1). Further, the display associated with the operation input shown in the second display example may be displayed according to the operation input of an external PC or the like (including operations other than the display).

  As shown in FIG. 8, when the display unit 202 receives the determination result of the importance level determination unit 37, the display unit 202 associates the continuous voice period of voice data (see FIG. 3), the recognized (estimated) work, and the work with time series. To display. In FIG. 8, the same reference numerals are given to the same operations as those shown in FIG.

  As shown in FIG. 8A, for example, the continuous sound periods d1 and d2 are associated with the period of the work A, and the continuous sound periods d3 to d6 are associated with the period of the work B. The display unit 202 displays the start time display 70 at the start time of the work A, and displays the end time display 72 at the end time of the work A. Further, the display unit 202 displays a start time display 74 at the start time of the work B, and displays an end time display 76 at the end time of the work B. The start time displays 70 and 74 and the end time displays 72 and 76 are slid by the pointer 60 according to an operation input through the operation receiving unit 40, for example, and the respective times can be changed back and forth. . That is, the voice recording selection apparatus 1 displays the start time and end time of the operations A and B by moving the start time displays 70 and 74 and the end time displays 72 and 76 via the operation receiving unit 40 by the user. And the time data can be changed.

  For example, in the period of the work B, the start time of the continuous sound period d3 is set as the start time of the work B, and the end time of the continuous sound period d6 is set as the end time of the work B. As shown in FIG. 8A, when the end time display 76 is slid on the display unit 202, the audio recording selection device 1 performs a process of changing the end time of the work B.

  Further, the end time of the work A does not coincide with the end time of the continuous sound period d2. That is, FIG. 8A shows a state in which the end time of the work A is changed as a result of the end time display 72 being moved regardless of the end time of the continuous audio period.

  FIG. 8B illustrates the operation of the audio recording selection device 1 when the end time display 76 is slid on the display unit 202. For example, when the end time display 76 is slid in the direction of moving the time forward and superimposed on the display of the continuous sound period d6, the sound recording selection device 1 starts the continuous sound period from the sound data recorded by the sound recording unit 31. The voice data issued at d6 is selected. And the audio | voice recording selection apparatus 1 performs the audio | voice output by the audio | voice output part 41 using the audio | voice data emitted in the continuous audio | voice period d6.

  In addition, when the end time display 76 is slid in the forward direction and overlapped with the operation B display, the audio recording selection device 1 starts from the audio data recorded by the audio recording unit 31 during the operation A period. You may be comprised so that the audio | voice data (audio | voice data of continuous audio | voice period d3-d6) emitted may be selected. And the audio | voice recording selection apparatus 1 performs the audio | voice output by the audio | voice output part 41 using the audio | voice data emitted in the continuous audio | voice periods d3-d6.

  The voice recording / selecting device 1 may be configured such that when the start time or the end time of the work is changed, the importance of the character string is determined again for the work whose start time or the end time is changed. . The display unit 202 may be configured to display the character string displayed in the first display example in addition to the information displayed in the second display example.

  Next, the operation of the audio recording selection device 1 will be described. FIG. 9 is a flowchart illustrating the operation of the audio recording selection device 1 according to the embodiment. In the audio recording / selecting device 1, for example, when the power is turned on, the audio receiving unit 30 starts to receive audio, and the information detecting unit 33 starts detection.

  As shown in FIG. 9, in step 100 (S100), the voice recognition unit 32 recognizes the voice received by the voice receiving unit 30, and converts it into a character string (text) associated with time.

  In step 102 (S102), the work recognition unit 35 accepts the acceleration data and the position data detected by the information detection unit 33, and recognizes the work (work information) associated with the user's time.

  In step 104 (S104), the importance level determination unit 37 uses the correspondence dictionary 36 to determine the importance level as a character string recognized by the voice recognition unit 32 in the work (work period) recognized by the work recognition unit 35. judge.

  In step 106 (S106), the display unit 202 accepts the result determined by the importance level determination unit 37, and displays the text separately according to the importance level.

  In this embodiment, the case where a specific work is recognized using the detection result of the information detection unit 33 and the importance of the character string recognized by the voice recognition unit 32 is determined with respect to the recognized work has been described as an example. However, the operation of the audio recording selection device 1 is not limited to this. For example, in the voice recording / selecting device 1, the work recognition unit 35 reads data (work information) such as a schedule showing a specific work having a start time and an end time, and a voice is sent to the work shown in the schedule. The importance determination unit 37 may be configured to determine the importance of the character string recognized by the recognition unit 32. The language in which the voice recording selection device 1 records voice and determines the importance of the character string is not limited to Japanese, and may be another language.

  According to the embodiment described above, the character string converted from the voice can be distinguished and output according to the importance in the specific work without taking the time equivalent to the recording time for recording the voice. .

  Moreover, although embodiment of this invention was described by the several combination, these embodiment is shown as an example and is not intending limiting the range of invention. These novel embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the spirit of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are included in the invention described in the claims and the equivalents thereof.

DESCRIPTION OF SYMBOLS 1 Voice recording selection apparatus 20 Operation part 200 Input part 202 Display part 21 Microphone 22 Detection part 220 Acceleration sensor 222 Positioning sensor 23 Speaker 24 Communication part 25 Storage part 26 Control part 30 Voice reception part 31 Voice recording part 32 Voice recognition part 33 Information Detection unit 330 Acceleration detection unit 332 Position detection unit 334 Pulse detection unit 336 Temperature detection unit 34 Corresponding model 35 Work recognition unit 36 Corresponding dictionary 37 Importance determination unit 38 Output unit 40 Operation reception unit 41 Audio output unit

Claims (9)

An audio receiving unit that receives audio from the user;
A voice recognition unit that recognizes the voice and converts it into a character string;
A work recognizing unit for recognizing work information indicating a specific work having a start time and an end time;
A storage for storing a correspondence dictionary in which at least one of a character string and a character string attribute important in the specific work and an importance level of the character string and the attribute in the specific work are associated with the specific work. And
Based on the work information and the correspondence dictionary, an importance level determination unit that determines the importance level of the character string converted from the speech accepted in the period from the disclosure time to the end time of the specific work,
An output unit that outputs a character string converted from the speech accepted in the period, according to the importance, and
An audio recording selection device having The voice recording selection device according to claim 1, wherein the work information indicates a plurality of specific works configured by a plurality of operations, and the importance of the attribute varies depending on the type of the work. A detector that detects at least one of a user's acceleration and the user's position;
The storage unit
Further storing a correspondence model indicating a correspondence between the specific task and at least one of a combination of movements specified by the user's acceleration and the user's position;
The work recognition unit
2. The work information is recognized according to at least one of a combination of movements identified by the user's acceleration detected by the detection unit, the position of the user detected by the detection unit, and the corresponding model. The audio recording selection device according to 1. The output unit is
The output for displaying the period according to the work information, and the output for distinguishing and displaying the character string converted from the voice received by the voice receiving unit during the period, according to the importance,
An operation receiving unit for receiving an input specifying the period displayed by the output unit;
An audio recording unit for recording the audio received by the audio receiving unit;
An audio output unit for outputting audio;
A control unit for controlling the audio output unit to output the audio recorded by the audio recording unit during the period specified by the input received by the operation receiving unit;
The voice recording / selecting device according to claim 3. The detector is
Further detecting at least one of the pulse of the user and the temperature of the environment where the user is located,
The corresponding model is
And further includes information indicating a correspondence between the specific operation and at least one of the pulse of the user detected by the detection unit and the temperature of the environment where the user is located,
The work recognition unit
According to the combination of movements identified by the user's acceleration detected by the detection unit, the position of the user, the pulse of the user, the temperature of the environment where the user is located, and the corresponding model The voice recording selection device according to claim 3, wherein the work information is recognized. The operation receiving unit is
Further accepting an input to change the period displayed by the output unit;
The controller is
The voice recording selection device according to claim 4, further controlling to change the work information in response to an input to change the period received by the operation receiving unit. The output unit is
Performing an output to further display a continuous audio period in which the audio recording unit records continuous audio within the period;
The operation receiving unit is
Further accepting input specifying the continuous speech period;
The controller is
The voice recording selection device according to claim 6, further controlling the voice output unit to output the voice recorded by the voice recording unit during the continuous voice period specified by the input received by the operation receiving unit. An audio recording selection method for selecting audio recording by an audio recording selection device ,
The voice recording selection device associates at least one of a character string and a character string attribute important in a specific work, and the importance of the character string and the attribute in the specific work with the specific work. A storage unit for storing a corresponding dictionary is provided.
A step of recognizing a voice from a user received by the voice receiving unit and converting the voice into a character string;
Recognizing work information indicating a specific work with a start time and an end time; and
Determining the importance of the character string converted from the speech accepted in the period from the disclosure time to the end time of the specific work based on the work information and the correspondence dictionary;
A step of distinguishing and outputting a character string converted from the speech accepted in the period according to the importance;
Voice recording selection method including. A step of recognizing a voice from a user received by the voice receiving unit and converting the voice into a character string;
Recognizing work information indicating a specific work for which a start time and an end time are defined; at least one of a character string and a character string attribute important in the specific work; the character string in the specific work; and Based on the correspondence dictionary in which the degree of importance of the attribute is associated with the specific work, and the work information, the voice is converted from the speech accepted in the period from the disclosure time to the end time of the specific work. Determining the importance of the string;
A character string converted from the speech accepted in the period, and output according to the importance,
Audio recording selection program for causing a computer to execute.

Images