Conversation aids

Abstract

To provide a conversation assistance device that can assist a speaker and a receiver who have a conversation with a transparent partition interposed therebetween.SOLUTION: A conversation assistance device 1 is a conversation assistance device that assists a speaker S and a receiver L who have a face-to-face conversation with a transparent partition 10 interposed therebetween, and includes: an utterance information acquisition unit 2 that acquires utterance content uttered by the speaker S as utterance information of a predetermined data format; an information processing unit 3 that subjects the utterance information to predetermined processing to generate received information of a predetermined output format; and an utterance information output unit 4 that outputs the received information to the receiver L in the predetermined output format.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a conversation assistance device. [Background technology]

[0002] Conventionally, as a device used in a situation where a speaker and a listener have a face-to-face conversation, a conversation assistance device that assists the speaker and the listener using an audio device such as a microphone and a speaker has been developed. For example, Patent Document 1 discloses an audio system that assists a customer and a counselor in a face-to-face conversation in a booth separated by a partition by combining a microphone on the counselor's side, a microphone on the customer's side, and a speaker located in the middle of both. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 5955340 Summary of the Invention [Problem to be solved by the invention]

[0004] Recently, in response to the spread of COVID-19, transparent partitions have been installed to separate people in various locations where face-to-face conversations take place, such as counters, accounting areas, and conference rooms, in order to prevent infection through droplets.

[0005] However, the audio system disclosed in Patent Document 1 was not designed to accommodate the installation of a transparent partition between the customer and the counselor, and had the problem that it could not adequately assist conversation between the two parties.

[0006] The present invention has been made in consideration of the above circumstances, and aims to provide a conversation assistance device that can assist a speaker and a listener having a conversation through a transparent partition. [Means for solving the problem]

[0007] The present invention is intended to solve the above-mentioned problems, and provides a conversation assistance device according to one embodiment of the present invention, A conversation assistance device that assists a speaker and a listener having a face-to-face conversation through a transparent partition, an utterance information acquisition unit that acquires the content of the utterance made by the speaker as utterance information in a predetermined data format; an information processing unit that performs a predetermined process on the speech information to generate received speech information in a predetermined output format; and a received call information output unit that outputs the received call information to the listener in the output format. Effect of the Invention

[0008] According to the conversation assistance device of the embodiment of the present invention, the speech information acquisition unit acquires the contents of a speaker's speech as speech information in a predetermined data format, the information processing unit performs a predetermined process on the speech information to generate received information in a predetermined output format, and the received information output unit outputs the received information to the listener. Therefore, it is possible to appropriately assist the speaker and the listener who are having a conversation through a transparent partition.

[0009] Other objects, configurations and effects will become apparent from the detailed description of the invention described below. [Brief description of the drawings]

[0010] [Figure 1] 1 is an overall configuration diagram showing an example of a conversation assistance device 1 according to an embodiment of the present invention. [Diagram 2] 1 is a block diagram showing an example of a conversation assistance device 1 according to an embodiment of the present invention. [Diagram 3] 1 is a block diagram showing an example of a conversation assistance device 1 according to a first embodiment of the present invention. [Figure 4] 13 is a graph showing acoustic propagation characteristics when various types of splash prevention garments are worn, compared to a reference case where no splash prevention garment is worn. [Diagram 5] FIG. 2 is a block diagram showing an example of a conversation assistance device 1 according to a second embodiment of the present invention. [Figure 6] FIG. 11 is a block diagram showing an example of a conversation assistance device 1 according to a third embodiment of the present invention. [Figure 7] FIG. 11 is a block diagram showing an example of a conversation assistance device 1 according to a fourth embodiment of the present invention. [Figure 8] FIG. 11 is a block diagram showing an example of a conversation assistance device 1 according to a fifth embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings.

[0012] (Embodiment) Fig. 1 is a diagram showing an overall configuration of an example of a conversation assistance device 1 according to an embodiment of the present invention. Fig. 2 is a block diagram showing an example of a conversation assistance device 1 according to an embodiment of the present invention.

[0013] The conversation assistance device 1 is used to assist a speaker S and a listener L who are having a face-to-face conversation through a transparent partition 10. The conversation assistance device 1 is applicable not only to cases where one person speaks as the speaker S and the other person listens as the listener L, but also to cases where the two people have a mutual conversation, i.e., cases where the positions of the speaker S and the listener L are alternated.

[0014] The partition 10 is placed between the speaker S and the listener L as a preventive measure against viral infections caused by droplets. The partition 10 is placed, for example, in an indoor space within a building, such as a counter, an accounting area, or a conference room. The partition 10 is configured like a screen that separates the speaker S and the listener L, and is placed on the floor or a desk. The partition 10 is made of, for example, a synthetic resin material such as acrylic or glass, and may be colorless and transparent, or colored and transparent. The size and shape of the partition 10 may be changed as appropriate depending on the place where the partition 10 is placed.

[0015] The conversation assistance device 1 mainly comprises an utterance information acquisition unit 2 that acquires the content of the utterance spoken by a speaker S as utterance information in a predetermined data format, an information processing unit 3 that performs predetermined processing (e.g., voice correction processing, voice recognition processing, image recognition processing, etc.) on the utterance information in the predetermined data format to generate received information in a predetermined output format, and a received information output unit 4 that outputs the received information to a listener L in the predetermined output format.

[0016] The speech information acquisition unit 2 is composed of, for example, a microphone 20 that acquires the speech content as speech information in a voice data format (voice utterance information), a camera 21 that acquires the speech content as speech information in an image data format (image utterance information), or an option input device 22 that acquires the speech content as speech information in an option data format (option utterance information), etc. The speech information acquisition unit 2 may include a plurality of each of the microphones 20, the cameras 21, and the option input devices 22, or may include a plurality of types of speech information acquisition units 2 (for example, the microphones 20 and the cameras 21, etc.) among the microphones 20, the cameras 21, and the option input devices 22.

[0017] The microphone 20 and the camera 21 may be, for example, those using the microphone function and the camera function of a mobile device (such as a smartphone) owned by the speaker S. The option input device 22 is composed of a non-contact touch device or the like that displays multiple options as characters, figures, etc. and can acquire the option selected by the speaker S. The option input device 22 may be a transparent panel integrated with the partition 10. The speech information (audio utterance information, image utterance information, option utterance information, etc.) in voice data format, image data format, option data format, etc. acquired by the speech information acquisition unit 2 is input to the information processing unit 3.

[0018] The received information output unit 4 is composed of, for example, a speaker 40 that outputs received information in a voice output format (voice received information) in a voice output format, a character display device 41 that outputs received information in a character output format (text received information) in a character output format, and an option output device 42 that outputs received information in an option output format (option received information) in the option output format, etc. The received information output unit 4 may include a plurality of each of the speaker 40, the character display device 41, and the option output device 42, or may include a plurality of types of received information output units 4 (for example, the speaker 40 and the character display device 41, etc.) among the speaker 40, the character display device 41, and the option output device 42.

[0019] The character display device 41 is composed of, for example, a liquid crystal display, an organic EL display, a projector capable of projecting an image onto the partition 10, or the like. The option output device 42 is composed of, for example, a non-contact touch device capable of displaying the option selected by the speaker S as characters or the like. Note that the character display device 41 and the option output device 42 may be made of a transparent panel integrated with the partition 10.

[0020] Conversation assistance device 1 may further include a speaker characteristic acquisition section 5 that acquires characteristic information of speaker S.

[0021] The speaker characteristic acquisition unit 5 is composed of, for example, a camera 50 that captures a speaker image including the face and upper body of the speaker S, and an operation device 51 that accepts operations of the speaker S using an operation mechanism such as a switch or button. The operation device 51 accepts speaker operation information that specifies the characteristics of the speaker S (age, presence or absence of hearing impairment, etc.) as characteristic information of the speaker S. The characteristic information of the speaker S acquired by the speaker characteristic acquisition unit 5 is input to the information processing unit 3. The camera 50 may be used in combination with the camera 21 constituting the utterance information acquisition unit 2. The operation device 51 may be used in combination with the option input device 22 constituting the utterance information acquisition unit 2.

[0022] The conversation assistance device 1 may further include a listener characteristic acquisition unit 6 that acquires listener L's characteristic information.

[0023] The listener characteristic acquisition unit 6 is composed of, for example, a camera 60 that captures a listener image including the face and upper body of the listener L, and an operation device 61 that accepts operations of the listener L using an operation mechanism such as a switch or a button. The operation device 61 accepts listener operation information that specifies the characteristics of the listener L (age, presence or absence of hearing impairment, etc.) as the listener L's characteristic information. The listener L's characteristic information acquired by the listener characteristic acquisition unit 6 is input to the information processing unit 3. The operation device 61 may also be used as the option output device 42 constituting the received information output unit 4.

[0024] The information processing unit 3 is configured, for example, by a general-purpose or dedicated computer. The information processing unit 3 includes an input circuit 30 that inputs speech information (speech speech information, image speech information, option speech information, etc.) in a voice data format, an image data format, an option data format, etc., a calculation processing unit 31 that performs a predetermined process on the speech information input to the input circuit 30 and generates received information (speech reception information, text reception information, option reception information, etc.) in a voice output format, a character output format, an option output format, etc., an output circuit 32 that outputs the received information generated by the calculation processing unit 31 to the received information output unit 4, and a storage unit 33 configured by a HDD, an SDD, a memory, etc. Note that the information processing unit 3 may be configured as a part of an audio device, a video device, etc.

[0025] The input circuit 30 and the output circuit 32 transmit and receive various information by wired or wireless communication, and any communication standard is applied. When the conversation assistance device 1 includes a speaker characteristic acquisition unit 5 and a listener characteristic acquisition unit 6, the input circuit 30 further receives characteristic information of the speaker S and characteristic information of the listener L. The information input to the input circuit 30 and the information output from the output circuit 32 may be either an analog signal or a digital signal.

[0026] The storage unit 33 stores a program 330 for controlling the operation of the arithmetic processing unit 31 and various setting information 331 to be referenced when the arithmetic processing unit 31 performs a predetermined process.

[0027] The arithmetic processing unit 31 is configured with one or more processors such as a CPU, an MPU, a GPU, etc., and performs predetermined processes such as audio processing, image processing, option discrimination processing, conversion processing, log storage processing, operation discrimination processing, etc. by executing a program 330 stored in the storage unit 33. The arithmetic processing unit 31 may be configured with hardware such as an FPGA or an ASIC, and may perform predetermined processes.

[0028] Examples of voice processing include a first or second voice correction process that amplifies voice in a first or second frequency band or removes noise from the voice utterance information based on characteristic information of the speaker S, a third voice correction process that amplifies voice in a third frequency band or removes noise from the voice utterance information based on characteristic information of the listener L, a voice speed adjustment process that speeds up or slows down the voice playback speed for the voice utterance information, and a voice recognition process that recognizes the speech content contained in the voice utterance information and converts it into text information.

[0029] Image processing includes, for example, face recognition processing that recognizes the face of a person included in the image speech information, speaker image, or listener image, and recognizes, for example, whether or not the person is wearing splash prevention equipment such as a mask or face shield, as well as their age, gender, etc.; action recognition processing that recognizes the actions of a person included in the image speech information, speaker image, or listener image; automatic lip reading processing that recognizes the speech content from the mouth movements (lip reading) of the speaker S included in the image speech information and converts it into text information; and sign language recognition processing that recognizes the speech content from the hand movements (sign language) of the speaker S included in the image speech information and converts it into text information.

[0030] The option discrimination process is, for example, a process of discriminating options included in option utterance information and converting the options into text information.

[0031] The conversion process is a process of converting character information (text information) acquired from speech information by voice processing or image processing into received information in a predetermined output format (e.g., voice output format, character output format, or option output format, etc.). The conversion process may further include at least one of an automatic translation process and a log storage process. The automatic translation process is a process of translating character information in a speech language (e.g., Japanese) used by the speaker S into character information in a received language (e.g., English) used by the listener L to generate received information. The log storage process is a process of generating received information including past speech contents by recording a predetermined amount or a predetermined time and storing the contents as speech information. The operation discrimination process is a process of discriminating the operation contents accepted by the operation device 51 and the operation device 61.

[0032] The conversation assistance device 1 can handle various combinations of the data format of the utterance information and the output format of the received information by arbitrarily combining the configurations of each of the utterance information acquisition unit 2, the received information output unit 4, the speaker characteristic acquisition unit 5, and the receiver characteristic acquisition unit 6. In this case, the information processing unit 3 performs a predetermined process according to the combination of the data format of the utterance information and the output format of the received information. In the following, in Examples 1 to 5, specific examples of combinations that can be realized by the conversation assistance device 1 will be described.

[0033] The conversation assistance device 1 can also be applied to a case where two people have a conversation, that is, where the positions of the speaker S and the listener L are alternately switched. In that case, the conversation assistance device 1 needs to handle a first situation (speaker S is the person on the left and listener L is the person on the right) as shown in FIG. 1 and a second situation (speaker S is the person on the right and listener L is the person on the left) in which the positions of the speaker S and listener L are switched with respect to FIG. 1. Therefore, the speech information acquisition unit 2, the received speech information output unit 4, the speaker characteristic acquisition unit 5, and the listener characteristic acquisition unit 6 are provided as a first set corresponding to the first situation and a second set corresponding to the second situation. In that case, the parts that can be used in common between the first set and the second set can be appropriately used in common, thereby simplifying the configuration of the entire device.

[0034] (First embodiment) FIG. 3 is a block diagram showing an example of a conversation assistance device 1 according to a first embodiment of the present invention.

[0035] A conversation assistance device 1 according to this embodiment includes a microphone 20 as an utterance information acquisition unit 2, an information processing unit 3, a speaker 40 as a received-talk information output unit 4, a camera 50 as a speaker characteristic acquisition unit 5, and a camera 60 and an operation device 61 as a receiver characteristic acquisition unit 6. Note that the basic configuration of each unit is the same as in the above embodiment, and therefore the following description will focus on the features of this embodiment.

[0036] The input circuit 30 receives voice speech information from the microphone 20, a speaker image from the camera 50, a listener image from the camera 60, and listener operation information from the operation device 61. The output circuit 32 outputs the voice reception information to the speaker 40.

[0037] The arithmetic processing unit 31 performs first voice correction processing, second voice correction processing, third voice correction processing, and voice speed adjustment processing on the voice utterance information from the microphone 20 to generate voice reception information. The arithmetic processing unit 31 performs face recognition processing to recognize whether the speaker S is wearing a splash prevention garment from the speaker image captured by the camera 50. The arithmetic processing unit 31 performs face recognition processing to recognize the age of the listener L from the listener image captured by the camera 60. The arithmetic processing unit 31 performs operation discrimination processing to recognize whether the listener L has a hearing impairment from the listener operation information from the operation device 61.

[0038] The arithmetic processing unit 31 performs the above processes in parallel or in a composite manner to assist the conversation between the speaker S and the listener L. Specifically, the arithmetic processing unit 31 performs face recognition processing on the speaker image to determine whether the speaker S is wearing a splash prevention garment as feature information of the speaker S.

[0039] As a result, when the calculation processing unit 31 determines that the speaker S is not wearing a splash prevention device, it applies a first audio correction process to the audio speech information input from the microphone 20, correcting the audio of the speaker S in a high frequency band (first frequency band) where the audio is attenuated due to the presence of the partition 10, and generates audio reception information (shown as audio information in Figure 1).

[0040] On the other hand, when the calculation processing unit 31 determines that the speaker S is wearing a splash prevention harness, it applies a second voice correction process to the voice speech information input from the microphone 20, correcting the voice in a frequency band (second frequency band) that is the combination of the high frequency band attenuated by the presence of the partition 10 and the high frequency band attenuated by wearing the splash prevention harness, and generates voice reception information.

[0041] Furthermore, when the processor 31 determines that the speaker S is wearing a splash prevention garment, it may further recognize the type of splash prevention garment. Examples of the types of splash prevention garments include pleated nonwoven masks, flat gauze masks, three-dimensional urethane masks, mouth shields, face shields with vertical shielding surfaces, and face shields with oblique shielding surfaces.

[0042] Fig. 4 is a graph showing acoustic propagation characteristics when various types of splash prevention garments are worn, with a reference being a case where no splash prevention garment is worn. In Fig. 4, the horizontal axis represents frequency, and the vertical axis represents relative sound pressure levels in 1 / 3 octave bands, and it can be seen that the frequency bands attenuated differ depending on the type of splash prevention garment. Therefore, when the calculation processing unit 31 uses the characteristics of such splash prevention garments to recognize the type of recognized splash prevention garment worn by the speaker S as characteristic information of the speaker S, the processing content of the second voice correction process (such as the range of high frequency bands to be amplified and the gain when amplifying) may be changed depending on the type of the recognized splash prevention garment.

[0043] In addition, the calculation processing unit 31 determines at least one of the age and the presence or absence of a hearing impairment of the listener L as characteristic information of the listener L by face recognition processing on the listener image or operation determination processing on the listener operation information.

[0044] As a result, when the calculation processing unit 31 determines that the listener L is elderly (age exceeds a predetermined reference age (e.g., 70 years old)) or has a hearing impairment, it applies a third voice correction process to the voice speech information (shown as voice information in Figure 1) input from the microphone 20, which amplifies a frequency band (third frequency band) including high frequency bands and frequency bands of consonant components that are difficult for elderly people and those with hearing impairments to hear, and corrects the voice to convert it into a voice that is easy for the listener L to hear, thereby generating voice reception information.

[0045] Next, the calculation processing unit 31 performs a voice speed adjustment process to slow down the voice playback speed on the voice utterance information (shown as voice information in FIG. 1) to generate voice reception information. On the other hand, if the calculation processing unit 31 determines that the listener L is not elderly and does not have a hearing impairment, it generates voice reception information by omitting the third voice correction process and the voice speed adjustment process. Note that the calculation processing unit 31 may perform at least one of the third voice correction process and the voice speed adjustment process based on the feature information of the listener L.

[0046] Then, the arithmetic processing unit 31 causes the speaker 40 to output (emit sound) the voice reception information generated as described above, so that the speech content of the speaker S is heard by the listener L as sound.

[0047] Although the conversation assistance device 1 according to the present embodiment has been described as including the camera 50 as the speaker feature acquisition unit 5, and the camera 60 and the operation device 61 as the receiver feature acquisition unit 6, it may also include either the speaker feature acquisition unit 5 or the receiver feature acquisition unit 6. In that case, if the conversation assistance device 1 does not include the speaker feature acquisition unit 5, it may perform either the first or second voice correction process. Also, if the conversation assistance device 1 does not include the speaker feature acquisition unit 5, it may perform the third voice correction process, or may not perform the third voice correction process.

[0048] As described above, according to this embodiment, the information processing unit 3 performs the first or second voice correction process depending on whether the speaker S is wearing a splash-proof wearing device. When the speaker S is not wearing a splash-proof wearing device, the voice of the speaker S is corrected taking into account the influence of the partition 10 and output from the speaker 40. When the speaker S is wearing a splash-proof wearing device, the voice of the speaker S is corrected taking into account not only the influence of the partition 10 but also the influence of the splash-proof wearing device and output from the speaker 40. In addition, the information processing unit 3 performs the third voice correction process and the voice speed adjustment process depending on the age and the presence or absence of hearing impairment of the listener L. For example, when the listener L is elderly or has a hearing impairment, the voice of the speaker S is corrected and the playback speed of the voice is adjusted and output from the speaker 40. Therefore, it is possible to assist smooth conversation while taking preventive measures against viral infections caused by droplets.

[0049] (Second Example) FIG. 5 is a block diagram showing an example of a conversation assistance device 1 according to a second embodiment of the present invention.

[0050] The conversation assistance device 1 according to this embodiment includes a microphone 20 as an utterance information acquisition unit 2, an information processing unit 3, and a character display device 41 as a received-talk information output unit 4. Note that the basic configuration of each unit is the same as in the above embodiment, so the following description will focus on the features of this embodiment.

[0051] Voice speech information is input to the input circuit 30 via the microphone 20. Text reception information is output from the output circuit 32 to the character display device 41.

[0052] The arithmetic processing unit 31 generates text information by performing voice recognition processing on the voice speech information from the microphone 20. The arithmetic processing unit 31 performs conversion processing to convert the text information into received text information.

[0053] The arithmetic processing unit 31 performs the above processes in parallel or in a composite manner to assist the conversation between the speaker S and the listener L. Specifically, the arithmetic processing unit 31 generates text information by performing voice recognition processing on the voice utterance information input from the microphone 20. Next, the arithmetic processing unit 31 generates text reception information by performing conversion processing on the text information. At this time, the arithmetic processing unit 31 may perform automatic translation processing on the text information. In addition, the arithmetic processing unit 31 may generate text reception information including the contents of past utterances by performing log storage processing on the text information. Then, the arithmetic processing unit 31 outputs (displays) the text reception information generated as described above on the character display device 41, so that the contents of the utterance of the speaker S are visually recognized by the listener L as text.

[0054] As described above, according to this embodiment, the information processing unit 3 converts the contents of the speaker S's speech into character information by speech recognition processing, and the character information is output to the character display device 41. Therefore, even in a situation where the listener L has difficulty hearing the contents of the speaker S's speech, the listener L can visually confirm the contents of the speaker S's speech on the character display device 41, thereby supporting smooth conversation while taking preventive measures against viral infections caused by droplets.

[0055] In addition, the information processing unit 3 can perform automatic translation processing to assist conversations even when there are differences in the languages ​​used. Furthermore, the information processing unit 3 can perform log storage processing to enable the listener L to refer to the contents of past utterances.

[0056] (Third Example) FIG. 6 is a block diagram showing an example of a conversation assistance device 1 according to a third embodiment of the present invention.

[0057] The conversation assistance device 1 according to this embodiment includes a camera 21 as an utterance information acquisition unit 2, an information processing unit 3, and a character display device 41 as a received-talk information output unit 4. Note that the basic configuration of each unit is the same as in the above embodiment, so the following description will focus on the features of this embodiment.

[0058] Image speech information from the camera 21 is input to the input circuit 30. Text speech information is output from the output circuit 32 to the character display device 41.

[0059] The arithmetic processing unit 31 generates text information by performing an automatic lip reading process on the image speech information from the camera 21. The arithmetic processing unit 31 performs a conversion process to convert the text information into received text information.

[0060] The arithmetic processing unit 31 performs the above processes in parallel or in a composite manner to assist the conversation between the speaker S and the listener L. Specifically, the arithmetic processing unit 31 generates text information by automatic lip reading processing of the image speech information input from the camera 21. Next, the arithmetic processing unit 31 generates text reception information by conversion processing of the text information. At this time, the arithmetic processing unit 31 may perform automatic translation processing on the text information. The arithmetic processing unit 31 may also generate text reception information including the contents of past utterances by performing log storage processing on the text information. Then, the arithmetic processing unit 31 outputs (displays) the text reception information generated as described above to the character display device 41, so that the contents of the utterance of the speaker S are visually recognized by the listener L as text.

[0061] As described above, according to this embodiment, the information processing unit 3 converts the speech content of the speaker S into text information by automatic lip reading processing, and the text information is output to the character display device 41. Therefore, even in a situation where the listener L has difficulty hearing the speech content of the speaker S, the listener L can visually confirm the speech content of the speaker S through the character display device 41, thereby supporting smooth conversation while taking preventive measures against viral infections caused by droplets.

[0062] (Fourth Example) FIG. 7 is a block diagram showing an example of a conversation assistance device 1 according to a fourth embodiment of the present invention.

[0063] The conversation assistance device 1 according to this embodiment includes a camera 21 as an utterance information acquisition unit 2, an information processing unit 3, and a character display device 41 as a received-talk information output unit 4. Note that the basic configuration of each unit is the same as in the above embodiment, so the following description will focus on the features of this embodiment.

[0064] Image speech information from the camera 21 is input to the input circuit 30. Text speech information is output from the output circuit 32 to the character display device 41.

[0065] The arithmetic processing unit 31 generates character information by performing sign language recognition processing on the image speech information from the camera 21. The arithmetic processing unit 31 performs conversion processing to convert the character information into received text information.

[0066] The arithmetic processing unit 31 performs the above processes in parallel or in a composite manner to assist the conversation between the speaker S and the listener L. Specifically, the arithmetic processing unit 31 generates text information by performing sign language recognition processing on the image speech information input from the camera 21. Next, the arithmetic processing unit 31 generates text reception information by performing conversion processing on the text information. At this time, the arithmetic processing unit 31 may perform automatic translation processing on the text information. Furthermore, the arithmetic processing unit 31 may generate text reception information including the contents of past utterances by performing log storage processing on the text information. Then, the arithmetic processing unit 31 outputs (displays) the text reception information generated as described above on the character display device 41, so that the contents of the utterance of the speaker S are visually recognized by the listener L as text.

[0067] As described above, according to this embodiment, the information processing unit 3 converts the contents of the speaker S's sign language utterance into character information by sign language recognition processing, and the character information is output to the character display device 41. Therefore, even in a situation where the listener L does not have sufficient knowledge of sign language, the contents of the speaker S's utterance can be visually recognized by the character display device 41, so that smooth conversation can be assisted while taking preventive measures against viral infections caused by droplets.

[0068] (Fifth Example) FIG. 8 is a block diagram showing an example of a conversation assistance device 1 according to a fifth embodiment of the present invention.

[0069] The conversation assistance device 1 according to the present embodiment includes an option input device 22 as an utterance information acquisition unit 2, an information processing unit 3, and an option output device 42 as a received-talk information output unit 4. Note that the basic configuration of each unit is the same as in the above embodiment, so the following description will focus on the features of the present embodiment.

[0070] The option speech information is input to the input circuit 30 by the option input device 22. The option speech information is output from the output circuit 32 to the option output device .

[0071] The arithmetic processing unit 31 generates character information by performing an option discrimination process on the option speech information from the option input device 22. The arithmetic processing unit 31 performs a conversion process to convert the character information into option speech information.

[0072] The calculation processing unit 31 performs the above processes in parallel or in a composite manner to assist the conversation between the speaker S and the listener L. Specifically, the calculation processing unit 31 generates character information by performing option discrimination processing on the option utterance information input from the option input device 22. Next, the calculation processing unit 31 generates option received information by performing conversion processing on the character information. Then, the calculation processing unit 31 outputs (displays) the option received information generated as described above to the option output device 42, so that the content of the utterance of the speaker S is visually recognized by the listener L as options (characters, symbols, etc.).

[0073] As described above, according to this embodiment, the information processing unit 3 converts the speech content of the speaker S according to the options into text information by the option discrimination process, and the text information is output to the option output device 42. Therefore, even in a situation where the listener L has difficulty hearing the speech content of the speaker S, the speech content of the speaker S can be visually confirmed by the option output device 42, so that smooth conversation can be assisted while taking preventive measures against viral infections caused by droplets.

[0074] (Other embodiments) Although the embodiments of the present invention (including Examples 1 to 5) have been described above, the present invention is not limited to the above-mentioned embodiments and can be modified as appropriate without departing from the technical spirit of the present invention.

[0075] In the above embodiment, in Examples 2 to 4, it has been described that the information processing unit 3 performs voice recognition processing on the utterance information to generate received information in a character output format, and the character display device 41 outputs the received information in the character output format. However, the information processing unit 3 may perform voice recognition processing on the utterance information to generate received information in a choice output format, and the choice output device 42 may output the received information in the choice output format.

[0076] In the above embodiment, each of the configurations of Examples 1 to 5 may be combined in plural, and in that case, the function realized by each configuration may be switched on and off.

[0077] In the above embodiment, the program 330 has been described as being stored in the storage unit 33, but it may be provided by being recorded in a computer-readable recording medium such as a CD-ROM or DVD as a file in an installable or executable format. Also, the program 330 may be provided by being stored on a computer connected to a network such as the Internet and downloaded via the network. [Explanation of symbols]

[0078] 1... conversation assistance device, 2... speech information acquisition unit, 3... information processing unit, 4... received call information output unit, 5... speaker characteristic acquisition unit, 6... listener characteristic acquisition unit, 10... partition, 20: microphone; 21: camera; 22: selection input device; 30: input circuit; 31: arithmetic processing unit; 32: output circuit; 33: storage unit; 40: speaker; 41: character display device; 42: option output device; 50: camera, 51: operation device, 60: camera, 61: operation device, 330... program, 331... setting information, L... receiver, S... speaker

Claims

[Claim 1] A conversation assistance device that assists a speaker and a listener having a face-to-face conversation through a transparent partition, an utterance information acquisition unit that acquires the content of the utterance made by the speaker as utterance information in a predetermined data format; an information processing unit that performs a predetermined process on the speech information to generate received speech information in a predetermined output format; a received information output unit that outputs the received information to the listener in the output format; a speaker feature acquisition unit that acquires feature information of the speaker by capturing an image of the speaker including a face of the speaker, The speech information acquisition unit a microphone for acquiring the speech content as the speech information in a voice data format; The information processing unit includes: determining whether the speaker is wearing a droplet prevention garment based on the speaker image; when it is determined that the droplet prevention wearing equipment is not being worn, a first voice correction process is performed on the speech information to correct a first frequency band, thereby generating the received speech information in a voice output format; when it is determined that the droplet prevention wearing gear is worn, a second voice correction process is performed on the speech information to correct a second frequency band, thereby generating the received speech information in a voice output format; The received call information output unit a speaker that outputs the received information in the audio output format; Conversation aid.

Images