System

Abstract

An object of the system according to the embodiment is to reproduce the voice and the way of speaking of the deceased and to reestablish emotional connection.SOLUTION: A system includes a collection unit, a learning unit, a reception unit, a generation unit, and a reproduction unit. The collection unit collects video and audio data of the deceased. The learning unit learns the data collected by the collection unit and analyzes the voice and the way of speaking of the deceased. The receiving unit receives an input from a user. The generation unit generates new words in the voice of the deceased based on the input accepted by the acceptance unit. The reproduction unit reproduces the words generated by the generation unit.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] Current technology offers limited means for maintaining an emotional connection with the deceased, leaving room for improvement.

[0005] The system of the embodiment aims to recreate the voice and speaking style of the deceased person and re-establish the emotional connection. [Means for solving the problem]

[0006] The system according to the embodiment includes a collection unit, a learning unit, a reception unit, a generation unit, and a playback unit. The collection unit collects video and audio data of the deceased. The learning unit learns the data collected by the collection unit and analyzes the voice and speaking style of the deceased. The reception unit receives user input. The generation unit generates new words in the voice of the deceased based on the input received by the reception unit. The playback unit plays back the words generated by the generation unit. [Effects of the Invention]

[0007] Embodiments of the system can recreate the voice and speaking style of the deceased person, allowing for the re-establishment of an emotional connection. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. DETAILED DESCRIPTION OF THE INVENTION

[0009] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0010] First, the terms used in the following description will be explained.

[0011] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, the processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), or a TPU (Tensor Processing Unit).

[0012] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0013] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0014] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), and Bluetooth (registered trademark).

[0015] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0016] [First embodiment] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0017] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0018] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0019] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0020] The reception device 38 includes a touch panel 38A and a microphone 38B, and receives user input. The touch panel 38A detects contact with a pointer (for example, a pen or a finger) to receive user input by the touch of the pointer. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 (see FIG. 2) acquires the data indicating the user input.

[0021] Output device 40 includes a display 40A and a speaker 40B, and presents data to a user by outputting the data in a form of expression that the user can perceive (e.g., audio and / or text). Display 40A displays visible information such as text and images in accordance with instructions from processor 46. Speaker 40B outputs audio in accordance with instructions from processor 46. Camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0022] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0023] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0024] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0025] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0026] In the smart device 14, the specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. The specific processing program 60 is used together with the specific processing program 56 by the data processing system 10. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific processing program 60 executed on the RAM 48. Note that the smart device 14 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the specific processing unit 290 using these models.

[0027] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device (e.g., a generation server) may have the data generation model 58. In this case, the data processing device 12 obtains a processing result (prediction result, etc.) using the data generation model 58 by communicating with the server device having the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device owned by a user (e.g., a mobile phone, a robot, a home appliance, etc.). Next, an example of processing by the data processing system 10 according to the first embodiment will be described.

[0028] (Example 1) A system according to an embodiment of the present invention generates and plays back new words in the voice of the deceased based on video and audio data of the deceased. This system allows family and friends to reconnect with the deceased and enjoy moments of healing. By recreating the voice and words of the deceased, the system allows them to feel closer to their memories and deepen their bond with the deceased. Furthermore, by recreating the voice and words of the deceased, it allows them to maintain their relationship with the deceased and reestablish an emotional connection.

[0029] The system according to the embodiment includes a collection unit, a learning unit, a reception unit, a generation unit, and a playback unit. The collection unit collects video and audio data of the deceased. For example, the collection unit can collect video clips and audio messages of the deceased speaking. The collection unit can also collect photographs and text data of the deceased. The learning unit learns from the data collected by the collection unit and analyzes the deceased's voice and speaking style. For example, the learning unit analyzes the deceased's vocal characteristics and speaking patterns and builds a model for generating new words based on the analysis. The learning unit can learn the deceased's vocal characteristics and speaking patterns using a generation AI (e.g., a text generation AI or a multimodal generation AI). The reception unit accepts user input. For example, the reception unit can input the content of the user's speech to the deceased. The reception unit can support various input methods, such as voice input, text input, and image input. The generation unit generates new words in the deceased's voice based on the input accepted by the reception unit. For example, the generation unit can generate natural-sounding responses in the deceased's voice based on the user's input. The generation unit can use a generation AI to generate new words in the voice of the deceased. The playback unit plays back the words generated by the generation unit. For example, the playback unit can play back a generated response in the voice of the deceased. The playback unit can select the optimal audio playback method taking into account the user's device information. As a result, the system according to the embodiment can generate and play back new words in the voice of the deceased based on video and audio data of the deceased.

[0030] The collection unit may collect video clips or audio messages of the deceased speaking. For example, the collection unit may collect video clips in MP4 format or audio messages in MP3 format. The collection unit may also collect audio data in WAV format. This allows for more accurate data to be provided by collecting video clips or audio messages of the deceased speaking.

[0031] The learning unit can analyze the vocal characteristics or speech patterns of the deceased and, based on that, build a model that generates new words. For example, the learning unit can analyze the tone and rhythm of the deceased's voice and, based on that, build a model that generates new words. The learning unit can also analyze the pitch of the deceased's voice and the pauses in their speech and, based on that, build a model that generates new words. Furthermore, the learning unit can analyze the deceased's unique expressions and phrases and, based on that, build a model that generates new words. In this way, by analyzing the vocal characteristics and speech patterns of the deceased, more natural words can be generated.

[0032] The reception unit allows the user to input what he or she would like to say to the deceased. For example, the reception unit allows the user to input a question such as "Good morning, how was your day today?" The reception unit also allows the user to input a message or instruction to the deceased. Furthermore, the reception unit can support various input methods, such as voice input, text input, and image input. This allows the user to reproduce communication with the deceased by inputting what he or she would like to say to the deceased.

[0033] The generation unit can generate an appropriate response in the voice of the deceased based on the user's input. For example, the generation unit can generate a natural response in the voice of the deceased to a question input by the user. The generation unit can also generate an appropriate response in the voice of the deceased to a message input by the user. Furthermore, the generation unit can also generate an appropriate response in the voice of the deceased to an instruction input by the user. In this way, a conversation with the deceased can be reproduced by generating a natural response based on the user's input.

[0034] The playback unit can play back the generated response. For example, the playback unit can play back the generated response in the voice of the deceased. The playback unit can also play back the generated response using an optimal audio playback method based on the user's device information. Furthermore, the playback unit can also play back the generated response at an optimal volume based on the user's ambient sound. In this way, by playing back the generated response in the deceased's voice, communication with the deceased can be recreated.

[0035] When collecting video and audio data of the deceased, the collection unit can evaluate the quality of the data and select appropriate data. For example, the collection unit can use AI to evaluate the resolution of the collected video data and the sound quality of the audio data, and select the highest quality data. The collection unit can also use AI to analyze the brightness of the video data and the noise level of the audio data, and select the optimal data. Furthermore, the collection unit can use AI to evaluate the frame rate of the video data and the sampling rate of the audio data, and select the optimal data. This allows the system to provide higher quality data by evaluating the quality of the data and selecting the optimal data.

[0036] The collection unit can adjust the frequency of data collection when collecting video and audio data of the deceased. For example, the collection unit can adjust the collection frequency using AI in response to a user request, allowing the necessary data to be collected efficiently. The collection unit can also automatically adjust the collection frequency using AI in response to the data collection status, thereby avoiding data duplication. Furthermore, the collection unit can also adjust the collection frequency using AI based on the user's usage status, allowing the necessary data to be collected in a timely manner. This allows the necessary data to be collected efficiently by adjusting the data collection frequency.

[0037] The collection unit can appropriately store the data when collecting video and audio data of the deceased. For example, the collection unit can store the video data in an optimal compression format to reduce the data volume. The collection unit can also store the audio data in an optimal format to reduce the data volume while maintaining the sound quality. Furthermore, the collection unit can integrate and store the video data and audio data to facilitate data management. This allows the data storage format to be optimized, reducing the data volume and facilitating management.

[0038] When collecting video and audio data of the deceased, the collection unit can prioritize collecting highly relevant data by taking into account the user's geographical location information. For example, if the user is in a specific area, the collection unit can prioritize collecting video and audio data of the deceased related to that area. Furthermore, if the user is traveling, the collection unit can prioritize collecting data of the deceased related to the travel destination. Furthermore, if the user is in the deceased's hometown, the collection unit can prioritize collecting data of the deceased related to that area. In this way, by prioritizing the collection of highly relevant data by taking into account the user's geographical location information, more relevant data can be provided.

[0039] When collecting video and audio data of the deceased, the collection unit can analyze the user's social media activities and collect related data. For example, the collection unit can collect video and audio data of the deceased that the user shared on social media. The collection unit can also analyze the content of the user's social media posts and collect related data of the deceased. Furthermore, the collection unit can collect related data of the deceased by referring to the activities of the user's friends on social media. In this way, by analyzing the user's social media activities and collecting related data, more relevant data can be provided.

[0040] When collecting video and audio data of the deceased, the collection unit can customize the collection method by reflecting the user's past feedback. The collection unit can customize the type of data to collect, for example, based on feedback provided by the user in the past. The collection unit can also adjust the priority of data to be collected by reflecting the user's past feedback. Furthermore, the collection unit can customize the collection frequency and storage format based on the user's past feedback. In this way, by customizing the collection method by reflecting the user's past feedback, more appropriate data can be collected.

[0041] During learning, the learning unit can analyze in detail the vocal characteristics and speaking patterns of the deceased to improve the accuracy of the learning model. For example, the learning unit can analyze in detail the pitch and tone of the deceased's voice and reflect this in the learning model. The learning unit can also analyze the rhythm and pauses of the deceased's speaking style and reflect this in the learning model. Furthermore, the learning unit can analyze the unique expressions and phrases of the deceased and reflect this in the learning model. In this way, the accuracy of the learning model can be improved by analyzing in detail the vocal characteristics and speaking patterns of the deceased.

[0042] During learning, the learning unit can optimize the learning algorithm by taking into account variations in the deceased's voice. For example, the learning unit can analyze variations in the volume of the deceased's voice and reflect them in the learning algorithm. The learning unit can also analyze variations in the pitch of the deceased's voice and reflect them in the learning algorithm. Furthermore, the learning unit can analyze variations in the tone of the deceased's voice and reflect them in the learning algorithm. In this way, by optimizing the learning algorithm by taking into account variations in the deceased's voice, a more natural voice can be generated.

[0043] The learning unit can remove background noise from the deceased's voice during learning and learn pure voice data. For example, the learning unit can remove noise in the background of the deceased's voice during learning and extract pure voice data. The learning unit can also remove music or environmental sounds in the background of the deceased's voice during learning and extract pure voice data. Furthermore, the learning unit can remove other people's voices in the background of the deceased's voice during learning and extract pure voice data. In this way, pure voice data can be learned by removing background noise from the deceased's voice.

[0044] During learning, the learning unit can analyze the video data of the deceased and link it to the voice characteristics. For example, the learning unit can analyze the mouth movements from the video data of the deceased and link it to the voice characteristics. The learning unit can also analyze facial expressions from the video data of the deceased and link it to the tone of voice and emotions. Furthermore, the learning unit can analyze gestures from the video data of the deceased and link it to the rhythm and pauses of the voice. In this way, by analyzing the video data of the deceased and linking it to the voice characteristics, a more natural voice can be generated.

[0045] During learning, the learning unit can analyze the deceased's handwritten notes and diary to complement the speaking pattern. For example, the learning unit can extract unique expressions and phrases from the deceased's handwritten notes to complement the speaking pattern. The learning unit can also extract emotional expressions from the deceased's diary to complement the speaking tone and rhythm. Furthermore, the learning unit can analyze context from the deceased's handwritten notes and diary to complement the speaking pattern. In this way, by analyzing the deceased's handwritten notes and diary, the speaking pattern can be complemented and a more natural voice can be generated.

[0046] During learning, the learning unit analyzes the tone and rhythm of the deceased's voice to generate a more natural voice. For example, the learning unit can analyze the tone of the deceased's voice in detail and reflect this in the learning model to generate a more natural voice. The learning unit can also analyze the rhythm of the deceased's voice and reflect this in the learning model to generate a more natural voice. Furthermore, the learning unit can analyze the pauses in the deceased's voice and reflect this in the learning model to generate a more natural voice. In this way, by analyzing the tone and rhythm of the deceased's voice, a more natural voice can be generated.

[0047] The reception unit can select the optimal reception means depending on the user's input method when receiving the input. For example, when the user inputs by voice, the reception unit can analyze the input content using voice recognition technology and provide appropriate feedback. Furthermore, when the user inputs by text, the reception unit can analyze the input content using natural language processing technology and provide appropriate feedback. Furthermore, when the user inputs by image, the reception unit can analyze the input content using image recognition technology and provide appropriate feedback. In this way, by selecting the optimal reception means depending on the user's input method, more appropriate feedback can be provided.

[0048] The reception unit can predict input contents by referring to the user's past input history when receiving the input. For example, the reception unit can predict what the user will input next based on what the user has previously input and present input candidates. The reception unit can also extract specific patterns from the user's past input history and predict input contents. Furthermore, the reception unit can analyze the user's past input history and predict and present frequently used phrases and words. In this way, by predicting input contents by referring to the user's past input history, more appropriate feedback can be provided.

[0049] The reception unit can analyze the user's input content in real time when receiving it and generate an appropriate response. The reception unit can, for example, analyze the user's input content in real time and immediately generate a response. The reception unit can also analyze the user's input content in real time and provide appropriate feedback. Furthermore, the reception unit can analyze the user's input content in real time and provide related information. In this way, by analyzing the user's input content in real time, an appropriate response can be immediately provided.

[0050] The reception unit can prioritize processing highly relevant input content during reception, taking into consideration the user's geographical location information. For example, if the user is in a specific area, the reception unit can prioritize processing input content related to that area. Furthermore, if the user is traveling, the reception unit can prioritize processing input content related to the travel destination. Furthermore, if the user is in the hometown of a deceased person, the reception unit can prioritize processing input content related to that area. In this way, by prioritizing highly relevant input content in consideration of the user's geographical location information, more appropriate feedback can be provided.

[0051] The reception unit can analyze the user's social media activity at the time of reception and suggest related input content. The reception unit can suggest related input content based on, for example, content shared by the user on social media. The reception unit can also analyze the content posted by the user on social media and suggest related input content. Furthermore, the reception unit can also suggest related input content by referring to the activity of the user's friends on social media. In this way, by analyzing the user's social media activity and suggesting related input content, it is possible to provide more appropriate feedback.

[0052] The reception unit can customize the reception method by reflecting the user's past feedback when receiving a request. The reception unit can customize the reception method, for example, based on feedback provided by the user in the past. The reception unit can also adjust the reception means by reflecting the user's past feedback. Furthermore, the reception unit can customize the reception interface based on the user's past feedback. In this way, by customizing the reception method by reflecting the user's past feedback, it is possible to provide more appropriate feedback.

[0053] During generation, the generation unit can generate more natural-sounding words by analyzing in detail the characteristics of the deceased's voice and speech patterns. For example, the generation unit can generate natural-sounding words by analyzing in detail the pitch and tone of the deceased's voice. The generation unit can also analyze the rhythm and pauses of the deceased's speech to generate natural-sounding words. Furthermore, the generation unit can analyze the deceased's unique expressions and phrases to generate natural-sounding words. In this way, by analyzing in detail the characteristics of the deceased's voice and speech patterns, more natural-sounding words can be generated.

[0054] The generation unit can adjust the tone and rhythm of the words to be generated based on the user's input content during generation. For example, if the user's input content is emotional, the generation unit can adjust the tone and rhythm to reflect the emotion. Furthermore, if the user's input content is a question, the generation unit can adjust the tone and rhythm to generate a natural response. Furthermore, if the user's input content is everyday conversation, the generation unit can adjust the tone and rhythm to generate a natural conversation. In this way, by adjusting the tone and rhythm of the words to be generated based on the user's input content, a more natural response can be generated.

[0055] During generation, the generation unit can reproduce background sounds of the deceased's voice to generate more realistic sound. The generation unit can, for example, reproduce environmental sounds that were in the background of the deceased's voice to generate realistic sound. The generation unit can also reproduce music that was in the background of the deceased's voice to generate realistic sound. Furthermore, the generation unit can also reproduce other people's voices that were in the background of the deceased's voice to generate realistic sound. In this way, by reproducing background sounds of the deceased's voice, more realistic sound can be generated.

[0056] During generation, the generation unit can analyze the video data of the deceased and link it to the voice characteristics. For example, the generation unit can analyze the mouth movements from the video data of the deceased and link it to the voice characteristics. The generation unit can also analyze facial expressions from the video data of the deceased and link it to the tone of voice and emotions. Furthermore, the generation unit can analyze gestures from the video data of the deceased and link it to the rhythm and pauses of the voice. In this way, by analyzing the video data of the deceased and linking it to the voice characteristics, a more natural voice can be generated.

[0057] During generation, the generation unit can analyze the deceased's handwritten notes or diary to complement the speaking pattern. For example, the generation unit can extract unique expressions and phrases from the deceased's handwritten notes to complement the speaking pattern. The generation unit can also extract emotional expressions from the deceased's diary to complement the speaking tone and rhythm. Furthermore, the generation unit can analyze context from the deceased's handwritten notes or diary to complement the speaking pattern. In this way, by analyzing the deceased's handwritten notes or diary, the speaking pattern can be complemented and a more natural voice can be generated.

[0058] During generation, the generation unit analyzes the tone and rhythm of the deceased's voice to generate a more natural voice. For example, the generation unit can analyze the tone of the deceased's voice in detail and reflect this in a learning model to generate a more natural voice. The generation unit can also analyze the rhythm of the deceased's voice and reflect this in a learning model to generate a more natural voice. Furthermore, the generation unit can analyze the pauses in the deceased's voice and reflect this in a learning model to generate a more natural voice. In this way, by analyzing the tone and rhythm of the deceased's voice, a more natural voice can be generated.

[0059] During playback, the playback unit can select the optimal audio playback method by taking into consideration the user's device information. For example, if the user is using a smartphone, the playback unit can select an audio playback method that matches the speaker characteristics of the device. Furthermore, if the user is using headphones, the playback unit can select an audio playback method that matches the headphone characteristics. Furthermore, if the user is using a smart speaker, the playback unit can select an audio playback method that matches the speaker characteristics. In this way, by selecting the optimal audio playback method by taking into consideration the user's device information, more appropriate audio can be played back.

[0060] During playback, the playback unit can customize the playback content by referring to the user's past playback history. The playback unit can customize the playback content, for example, based on content that the user has played in the past. The playback unit can also extract specific patterns from the user's past playback history and customize the playback content. Furthermore, the playback unit can analyze the user's past playback history and prioritize playback of frequently played content. In this way, by customizing the playback content by referring to the user's past playback history, more appropriate audio can be played.

[0061] During playback, the playback unit can analyze the user's current environmental sound and play back at an optimal volume. The playback unit can, for example, analyze the environmental sound around the user and play back audio at an optimal volume. Furthermore, the playback unit can also play back audio at a lower volume when the user is in a quiet environment. Furthermore, the playback unit can also play back audio at a higher volume when the user is in a noisy environment. In this way, by analyzing the user's current environmental sound and playing back at an optimal volume, more appropriate audio can be played back.

[0062] During playback, the playback unit can prioritize playback of highly relevant audio in consideration of the user's geographical location information. For example, if the user is in a specific area, the playback unit can prioritize playback of audio related to that area. Furthermore, if the user is traveling, the playback unit can prioritize playback of audio related to the travel destination. Furthermore, if the user is in the hometown of a deceased person, the playback unit can prioritize playback of audio related to that area. In this way, by prioritizing playback of highly relevant audio in consideration of the user's geographical location information, more appropriate audio can be played.

[0063] During playback, the playback unit can analyze the user's social media activity and play back related audio. For example, the playback unit can play back related audio based on content shared by the user on social media. The playback unit can also analyze the content posted by the user on social media and play back related audio. Furthermore, the playback unit can also play back related audio with reference to the activity of the user's friends on social media. In this way, by analyzing the user's social media activity and playing back related audio, more appropriate audio can be played back.

[0064] The playback unit can customize the playback method by reflecting the user's past feedback during playback. The playback unit can customize the playback method, for example, based on feedback provided by the user in the past. The playback unit can also adjust the playback means by reflecting the user's past feedback. Furthermore, the playback unit can customize the playback interface based on the user's past feedback. In this way, by customizing the playback method by reflecting the user's past feedback, more appropriate audio can be played back.

[0065] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0066] When collecting video and audio data of the deceased, the collection unit can diversify the sources of data collection. For example, the collection unit can collect video and audio data posted from the deceased's social media accounts. The collection unit can also collect data from archives of television and radio programs in which the deceased appeared. Furthermore, the collection unit can collect recordings of events and lectures in which the deceased participated. This allows for the collection of more diverse data and a wealth of material for recreating the deceased's voice and video.

[0067] The reception unit can adjust the tone and rhythm of the words generated in the voice of the deceased based on the content input by the user. For example, if the content input by the user is emotional, the reception unit can instruct the generation unit to generate words with a tone and rhythm that reflects the emotion. Also, if the content input by the user is a question, the reception unit can instruct the generation unit to generate a natural response. Furthermore, if the content input by the user is everyday conversation, the reception unit can instruct the generation unit to generate natural conversation. This makes it possible to generate appropriate words according to the content input by the user.

[0068] The generation unit can customize the content of the words generated in the voice of the deceased based on the user's input. For example, if the user speaks about a specific theme, the generation unit can generate words related to that theme. Also, if the user speaks about a specific event, the generation unit can generate words related to that event. Furthermore, if the user expresses a specific emotion, the generation unit can generate words corresponding to that emotion. In this way, customized words can be generated according to the user's input.

[0069] The system also includes a data analysis unit that can analyze the collected data and extract specific characteristics and patterns of the deceased. For example, the data analysis unit can extract phrases and expressions frequently used by the deceased. The data analysis unit can also analyze the words the deceased used in specific situations. Furthermore, the data analysis unit can analyze changes in the tone and rhythm of the deceased's voice and build a model to generate new words based on that analysis. This allows the system to generate more natural-sounding words that reflect the characteristics and patterns of the deceased.

[0070] The collection unit may diversify the data collection methods when collecting video and audio data of the deceased. For example, the collection unit may collect video and audio data provided by the deceased's family and friends. The collection unit may also collect data directly from devices used by the deceased. Furthermore, the collection unit may collect data from online communities and forums in which the deceased participated. This allows for the collection of more diverse data and enriches the material for recreating the deceased's voice and video.

[0071] The processing flow of the first embodiment will be briefly explained below.

[0072] Step 1: The collection unit collects video and audio data of the deceased. For example, the collection unit can collect video clips and audio messages of the deceased speaking. The collection unit can also collect photographs and text data of the deceased. Step 2: The learning unit learns from the data collected by the collection unit and analyzes the deceased's voice and speaking style. For example, the learning unit analyzes the deceased's voice characteristics and speaking patterns and builds a model to generate new words based on them. The learning unit can use generative AI (e.g., text generation AI or multimodal generation AI) to learn the deceased's voice characteristics and speaking patterns. Step 3: The reception unit receives input from the user. For example, the reception unit can input what the user will say to the deceased. The reception unit can support various input methods, such as voice input, text input, and image input. Step 4: The generation unit generates new words in the voice of the deceased based on the input received by the reception unit. For example, the generation unit can generate natural responses in the voice of the deceased based on the user's input. The generation unit can use a generation AI to generate new words in the voice of the deceased. Step 5: The playback unit plays back the words generated by the generation unit. For example, the playback unit can play back the generated response in the deceased's voice. The playback unit can select the optimal voice playback method taking into account the user's device information.

[0073] (Example 2) A system according to an embodiment of the present invention generates and plays back new words in the voice of the deceased based on video and audio data of the deceased. This system allows family and friends to reconnect with the deceased and enjoy moments of healing. By recreating the voice and words of the deceased, the system allows them to feel closer to their memories and deepen their bond with the deceased. Furthermore, by recreating the voice and words of the deceased, it allows them to maintain their relationship with the deceased and reestablish an emotional connection.

[0074] The system according to the embodiment includes a collection unit, a learning unit, a reception unit, a generation unit, and a playback unit. The collection unit collects video and audio data of the deceased. For example, the collection unit can collect video clips and audio messages of the deceased speaking. The collection unit can also collect photographs and text data of the deceased. The learning unit learns from the data collected by the collection unit and analyzes the deceased's voice and speaking style. For example, the learning unit analyzes the deceased's vocal characteristics and speaking patterns and builds a model for generating new words based on the analysis. The learning unit can learn the deceased's vocal characteristics and speaking patterns using a generation AI (e.g., a text generation AI or a multimodal generation AI). The reception unit accepts user input. For example, the reception unit can input the content of the user's speech to the deceased. The reception unit can support various input methods, such as voice input, text input, and image input. The generation unit generates new words in the deceased's voice based on the input accepted by the reception unit. For example, the generation unit can generate natural-sounding responses in the deceased's voice based on the user's input. The generation unit can use a generation AI to generate new words in the voice of the deceased. The playback unit plays back the words generated by the generation unit. For example, the playback unit can play back a generated response in the voice of the deceased. The playback unit can select the optimal audio playback method taking into account the user's device information. As a result, the system according to the embodiment can generate and play back new words in the voice of the deceased based on video and audio data of the deceased.

[0075] The collection unit may collect video clips or audio messages of the deceased speaking. For example, the collection unit may collect video clips in MP4 format or audio messages in MP3 format. The collection unit may also collect audio data in WAV format. This allows for more accurate data to be provided by collecting video clips or audio messages of the deceased speaking.

[0076] The learning unit can analyze the vocal characteristics or speech patterns of the deceased and, based on that, build a model that generates new words. For example, the learning unit can analyze the tone and rhythm of the deceased's voice and, based on that, build a model that generates new words. The learning unit can also analyze the pitch of the deceased's voice and the pauses in their speech and, based on that, build a model that generates new words. Furthermore, the learning unit can analyze the deceased's unique expressions and phrases and, based on that, build a model that generates new words. In this way, by analyzing the vocal characteristics and speech patterns of the deceased, more natural words can be generated.

[0077] The reception unit allows the user to input what he or she would like to say to the deceased. For example, the reception unit allows the user to input a question such as "Good morning, how was your day today?" The reception unit also allows the user to input a message or instruction to the deceased. Furthermore, the reception unit can support various input methods, such as voice input, text input, and image input. This allows the user to reproduce communication with the deceased by inputting what he or she would like to say to the deceased.

[0078] The generation unit can generate an appropriate response in the voice of the deceased based on the user's input. For example, the generation unit can generate a natural response in the voice of the deceased to a question input by the user. The generation unit can also generate an appropriate response in the voice of the deceased to a message input by the user. Furthermore, the generation unit can also generate an appropriate response in the voice of the deceased to an instruction input by the user. In this way, a conversation with the deceased can be reproduced by generating a natural response based on the user's input.

[0079] The playback unit can play back the generated response. For example, the playback unit can play back the generated response in the voice of the deceased. The playback unit can also play back the generated response using an optimal audio playback method based on the user's device information. Furthermore, the playback unit can also play back the generated response at an optimal volume based on the user's ambient sound. In this way, by playing back the generated response in the deceased's voice, communication with the deceased can be recreated.

[0080] The system can estimate the user's emotions and adjust the type of data to be collected based on the estimated user emotions. For example, if the user is sad, the emotion engine can estimate and prioritize the collection of video and audio data filled with happy memories of the deceased. Alternatively, if the user is feeling nostalgic, the emotion engine can estimate and collect video and audio data of the deceased when they were young. Furthermore, if the user is relaxed, the emotion engine can estimate and collect data of the deceased's everyday conversations and peaceful scenes. This allows for the collection of more appropriate data by adjusting the type of data to be collected based on the user's emotions.

[0081] When collecting video and audio data of the deceased, the collection unit can evaluate the quality of the data and select appropriate data. For example, the collection unit can use AI to evaluate the resolution of the collected video data and the sound quality of the audio data, and select the highest quality data. The collection unit can also use AI to analyze the brightness of the video data and the noise level of the audio data, and select the optimal data. Furthermore, the collection unit can use AI to evaluate the frame rate of the video data and the sampling rate of the audio data, and select the optimal data. This allows the system to provide higher quality data by evaluating the quality of the data and selecting the optimal data.

[0082] The collection unit can adjust the frequency of data collection when collecting video and audio data of the deceased. For example, the collection unit can adjust the collection frequency using AI in response to a user request, allowing the necessary data to be collected efficiently. The collection unit can also automatically adjust the collection frequency using AI in response to the data collection status, thereby avoiding data duplication. Furthermore, the collection unit can also adjust the collection frequency using AI based on the user's usage status, allowing the necessary data to be collected in a timely manner. This allows the necessary data to be collected efficiently by adjusting the data collection frequency.

[0083] The collection unit can appropriately store the data when collecting video and audio data of the deceased. For example, the collection unit can store the video data in an optimal compression format to reduce the data volume. The collection unit can also store the audio data in an optimal format to reduce the data volume while maintaining the sound quality. Furthermore, the collection unit can integrate and store the video data and audio data to facilitate data management. This allows the data storage format to be optimized, reducing the data volume and facilitating management.

[0084] The collection unit can estimate the user's emotions and determine the priority of data to be collected based on the estimated user emotions. For example, if the user is sad, the emotion engine can estimate and prioritize collecting data filled with happy memories of the deceased. Also, if the user is feeling nostalgic, the emotion engine can estimate and prioritize collecting data of the deceased's younger days. Furthermore, if the user is relaxed, the emotion engine can estimate and prioritize collecting data of the deceased's everyday conversations and peaceful scenes. In this way, by determining the priority of data to be collected based on the user's emotions, more appropriate data can be collected preferentially.

[0085] When collecting video and audio data of the deceased, the collection unit can prioritize collecting highly relevant data by taking into account the user's geographical location information. For example, if the user is in a specific area, the collection unit can prioritize collecting video and audio data of the deceased related to that area. Furthermore, if the user is traveling, the collection unit can prioritize collecting data of the deceased related to the travel destination. Furthermore, if the user is in the deceased's hometown, the collection unit can prioritize collecting data of the deceased related to that area. In this way, by prioritizing the collection of highly relevant data by taking into account the user's geographical location information, more relevant data can be provided.

[0086] When collecting video and audio data of the deceased, the collection unit can analyze the user's social media activities and collect related data. For example, the collection unit can collect video and audio data of the deceased that the user shared on social media. The collection unit can also analyze the content of the user's social media posts and collect related data of the deceased. Furthermore, the collection unit can collect related data of the deceased by referring to the activities of the user's friends on social media. In this way, by analyzing the user's social media activities and collecting related data, more relevant data can be provided.

[0087] When collecting video and audio data of the deceased, the collection unit can customize the collection method by reflecting the user's past feedback. The collection unit can customize the type of data to collect, for example, based on feedback provided by the user in the past. The collection unit can also adjust the priority of data to be collected by reflecting the user's past feedback. Furthermore, the collection unit can customize the collection frequency and storage format based on the user's past feedback. In this way, by customizing the collection method by reflecting the user's past feedback, more appropriate data can be collected.

[0088] The learning unit can estimate the user's emotions and select learning data based on the estimated user emotions. For example, if the user is sad, the learning unit can estimate using the emotion engine and select data filled with happy memories of the deceased as learning data. In addition, if the user is feeling nostalgic, the learning unit can estimate using the emotion engine and select data of the deceased when they were young as learning data. Furthermore, if the user is relaxed, the learning unit can estimate using the emotion engine and select data of the deceased's everyday conversations or calm scenes as learning data. In this way, by selecting learning data based on the user's emotions, more appropriate data can be learned.

[0089] During learning, the learning unit can analyze in detail the vocal characteristics and speaking patterns of the deceased to improve the accuracy of the learning model. For example, the learning unit can analyze in detail the pitch and tone of the deceased's voice and reflect this in the learning model. The learning unit can also analyze the rhythm and pauses of the deceased's speaking style and reflect this in the learning model. Furthermore, the learning unit can analyze the unique expressions and phrases of the deceased and reflect this in the learning model. In this way, the accuracy of the learning model can be improved by analyzing in detail the vocal characteristics and speaking patterns of the deceased.

[0090] During learning, the learning unit can optimize the learning algorithm by taking into account variations in the deceased's voice. For example, the learning unit can analyze variations in the volume of the deceased's voice and reflect them in the learning algorithm. The learning unit can also analyze variations in the pitch of the deceased's voice and reflect them in the learning algorithm. Furthermore, the learning unit can analyze variations in the tone of the deceased's voice and reflect them in the learning algorithm. In this way, by optimizing the learning algorithm by taking into account variations in the deceased's voice, a more natural voice can be generated.

[0091] The learning unit can remove background noise from the deceased's voice during learning and learn pure voice data. For example, the learning unit can remove noise in the background of the deceased's voice during learning and extract pure voice data. The learning unit can also remove music or environmental sounds in the background of the deceased's voice during learning and extract pure voice data. Furthermore, the learning unit can remove other people's voices in the background of the deceased's voice during learning and extract pure voice data. In this way, pure voice data can be learned by removing background noise from the deceased's voice.

[0092] The learning unit can estimate the user's emotions and adjust the learning frequency based on the estimated user's emotions. For example, if the user is sad, the learning unit can estimate the emotion engine and increase the learning frequency to quickly generate new words. Also, if the user is feeling nostalgic, the learning unit can estimate the emotion engine and adjust the learning frequency to generate new words at an appropriate time. Furthermore, if the user is relaxed, the learning unit can estimate the emotion engine and decrease the learning frequency to slowly generate new words. In this way, by adjusting the learning frequency based on the user's emotions, learning can be performed at a more appropriate time.

[0093] During learning, the learning unit can analyze the video data of the deceased and link it to the voice characteristics. For example, the learning unit can analyze the mouth movements from the video data of the deceased and link it to the voice characteristics. The learning unit can also analyze facial expressions from the video data of the deceased and link it to the tone of voice and emotions. Furthermore, the learning unit can analyze gestures from the video data of the deceased and link it to the rhythm and pauses of the voice. In this way, by analyzing the video data of the deceased and linking it to the voice characteristics, a more natural voice can be generated.

[0094] During learning, the learning unit can analyze the deceased's handwritten notes and diary to complement the speaking pattern. For example, the learning unit can extract unique expressions and phrases from the deceased's handwritten notes to complement the speaking pattern. The learning unit can also extract emotional expressions from the deceased's diary to complement the speaking tone and rhythm. Furthermore, the learning unit can analyze context from the deceased's handwritten notes and diary to complement the speaking pattern. In this way, by analyzing the deceased's handwritten notes and diary, the speaking pattern can be complemented and a more natural voice can be generated.

[0095] During learning, the learning unit analyzes the tone and rhythm of the deceased's voice to generate a more natural voice. For example, the learning unit can analyze the tone of the deceased's voice in detail and reflect this in the learning model to generate a more natural voice. The learning unit can also analyze the rhythm of the deceased's voice and reflect this in the learning model to generate a more natural voice. Furthermore, the learning unit can analyze the pauses in the deceased's voice and reflect this in the learning model to generate a more natural voice. In this way, by analyzing the tone and rhythm of the deceased's voice, a more natural voice can be generated.

[0096] The reception unit can estimate the user's emotion and provide feedback on the input content based on the estimated user's emotion. For example, if the user is sad, the emotion engine can estimate it and provide feedback in kind words. Also, if the user is feeling nostalgic, the reception unit can estimate it and provide feedback related to memories. Furthermore, if the user is relaxed, the emotion engine can estimate it and provide feedback in kind words. In this way, by providing feedback on the input content based on the user's emotion, more appropriate feedback can be provided.

[0097] The reception unit can select the optimal reception means depending on the user's input method when receiving the input. For example, when the user inputs by voice, the reception unit can analyze the input content using voice recognition technology and provide appropriate feedback. Furthermore, when the user inputs by text, the reception unit can analyze the input content using natural language processing technology and provide appropriate feedback. Furthermore, when the user inputs by image, the reception unit can analyze the input content using image recognition technology and provide appropriate feedback. In this way, by selecting the optimal reception means depending on the user's input method, more appropriate feedback can be provided.

[0098] The reception unit can predict input contents by referring to the user's past input history when receiving the input. For example, the reception unit can predict what the user will input next based on what the user has previously input and present input candidates. The reception unit can also extract specific patterns from the user's past input history and predict input contents. Furthermore, the reception unit can analyze the user's past input history and predict and present frequently used phrases and words. In this way, by predicting input contents by referring to the user's past input history, more appropriate feedback can be provided.

[0099] The reception unit can analyze the user's input content in real time when receiving it and generate an appropriate response. The reception unit can, for example, analyze the user's input content in real time and immediately generate a response. The reception unit can also analyze the user's input content in real time and provide appropriate feedback. Furthermore, the reception unit can analyze the user's input content in real time and provide related information. In this way, by analyzing the user's input content in real time, an appropriate response can be immediately provided.

[0100] The reception unit can estimate the user's emotion and determine the priority of input content based on the estimated user's emotion. For example, if the user is sad, the emotion engine can estimate that the reception unit is sad and prioritize processing input content that requires emotional support. The reception unit can also estimate that the emotion engine is nostalgic and prioritize processing input content related to memories. Furthermore, the reception unit can estimate that the emotion engine is relaxed and prioritize processing calm input content. In this way, by determining the priority of input content based on the user's emotion, the input content can be processed in a more appropriate order.

[0101] The reception unit can prioritize processing highly relevant input content during reception, taking into consideration the user's geographical location information. For example, if the user is in a specific area, the reception unit can prioritize processing input content related to that area. Furthermore, if the user is traveling, the reception unit can prioritize processing input content related to the travel destination. Furthermore, if the user is in the hometown of a deceased person, the reception unit can prioritize processing input content related to that area. In this way, by prioritizing highly relevant input content in consideration of the user's geographical location information, more appropriate feedback can be provided.

[0102] The reception unit can analyze the user's social media activity at the time of reception and suggest related input content. The reception unit can suggest related input content based on, for example, content shared by the user on social media. The reception unit can also analyze the content posted by the user on social media and suggest related input content. Furthermore, the reception unit can also suggest related input content by referring to the activity of the user's friends on social media. In this way, by analyzing the user's social media activity and suggesting related input content, it is possible to provide more appropriate feedback.

[0103] The reception unit can customize the reception method by reflecting the user's past feedback when receiving a request. The reception unit can customize the reception method, for example, based on feedback provided by the user in the past. The reception unit can also adjust the reception means by reflecting the user's past feedback. Furthermore, the reception unit can customize the reception interface based on the user's past feedback. In this way, by customizing the reception method by reflecting the user's past feedback, it is possible to provide more appropriate feedback.

[0104] The generation unit can estimate the user's emotion and adjust the expression method of the generated words based on the estimated user's emotion. For example, if the user is sad, the generation unit can use the emotion engine to estimate and generate a reply using kind words. Also, if the user is feeling nostalgic, the generation unit can use the emotion engine to estimate and generate a reply using words related to memories. Furthermore, if the user is relaxed, the generation unit can use the emotion engine to estimate and generate a reply using gentle words. In this way, by adjusting the expression method of the generated words based on the user's emotion, more appropriate words can be generated.

[0105] During generation, the generation unit can generate more natural-sounding words by analyzing in detail the characteristics of the deceased's voice and speech patterns. For example, the generation unit can generate natural-sounding words by analyzing in detail the pitch and tone of the deceased's voice. The generation unit can also analyze the rhythm and pauses of the deceased's speech to generate natural-sounding words. Furthermore, the generation unit can analyze the deceased's unique expressions and phrases to generate natural-sounding words. In this way, by analyzing in detail the characteristics of the deceased's voice and speech patterns, more natural-sounding words can be generated.

[0106] The generation unit can adjust the tone and rhythm of the words to be generated based on the user's input content during generation. For example, if the user's input content is emotional, the generation unit can adjust the tone and rhythm to reflect the emotion. Furthermore, if the user's input content is a question, the generation unit can adjust the tone and rhythm to generate a natural response. Furthermore, if the user's input content is everyday conversation, the generation unit can adjust the tone and rhythm to generate a natural conversation. In this way, by adjusting the tone and rhythm of the words to be generated based on the user's input content, a more natural response can be generated.

[0107] During generation, the generation unit can reproduce background sounds of the deceased's voice to generate more realistic sound. The generation unit can, for example, reproduce environmental sounds that were in the background of the deceased's voice to generate realistic sound. The generation unit can also reproduce music that was in the background of the deceased's voice to generate realistic sound. Furthermore, the generation unit can also reproduce other people's voices that were in the background of the deceased's voice to generate realistic sound. In this way, by reproducing background sounds of the deceased's voice, more realistic sound can be generated.

[0108] The generation unit can estimate the user's emotion and adjust the length of the generated words based on the estimated user emotion. For example, if the user is sad, the generation unit can estimate using the emotion engine to generate short, gentle words. Also, if the user is feeling nostalgic, the generation unit can estimate using the emotion engine to generate longer words related to memories. Furthermore, if the user is relaxed, the generation unit can estimate using the emotion engine to generate gentle, longer words. In this way, by adjusting the length of the generated words based on the user's emotion, more appropriate words can be generated.

[0109] During generation, the generation unit can analyze the video data of the deceased and link it to the voice characteristics. For example, the generation unit can analyze the mouth movements from the video data of the deceased and link it to the voice characteristics. The generation unit can also analyze facial expressions from the video data of the deceased and link it to the tone of voice and emotions. Furthermore, the generation unit can analyze gestures from the video data of the deceased and link it to the rhythm and pauses of the voice. In this way, by analyzing the video data of the deceased and linking it to the voice characteristics, a more natural voice can be generated.

[0110] During generation, the generation unit can analyze the deceased's handwritten notes or diary to complement the speaking pattern. For example, the generation unit can extract unique expressions and phrases from the deceased's handwritten notes to complement the speaking pattern. The generation unit can also extract emotional expressions from the deceased's diary to complement the speaking tone and rhythm. Furthermore, the generation unit can analyze context from the deceased's handwritten notes or diary to complement the speaking pattern. In this way, by analyzing the deceased's handwritten notes or diary, the speaking pattern can be complemented and a more natural voice can be generated.

[0111] During generation, the generation unit analyzes the tone and rhythm of the deceased's voice to generate a more natural voice. For example, the generation unit can analyze the tone of the deceased's voice in detail and reflect this in a learning model to generate a more natural voice. The generation unit can also analyze the rhythm of the deceased's voice and reflect this in a learning model to generate a more natural voice. Furthermore, the generation unit can analyze the pauses in the deceased's voice and reflect this in a learning model to generate a more natural voice. In this way, by analyzing the tone and rhythm of the deceased's voice, a more natural voice can be generated.

[0112] The playback unit can estimate the user's emotion and adjust the tone of the audio to be played back based on the estimated user's emotion. For example, if the user is sad, the emotion engine can estimate it and play back audio in a gentle tone. Also, if the user is feeling nostalgic, the emotion engine can estimate it and play back audio in a tone associated with memories. Furthermore, if the user is relaxed, the emotion engine can estimate it and play back audio in a gentle tone. In this way, by adjusting the tone of the audio to be played back based on the user's emotion, more appropriate audio can be played back.

[0113] During playback, the playback unit can select the optimal audio playback method by taking into consideration the user's device information. For example, if the user is using a smartphone, the playback unit can select an audio playback method that matches the speaker characteristics of the device. Furthermore, if the user is using headphones, the playback unit can select an audio playback method that matches the headphone characteristics. Furthermore, if the user is using a smart speaker, the playback unit can select an audio playback method that matches the speaker characteristics. In this way, by selecting the optimal audio playback method by taking into consideration the user's device information, more appropriate audio can be played back.

[0114] During playback, the playback unit can customize the playback content by referring to the user's past playback history. The playback unit can customize the playback content, for example, based on content that the user has played in the past. The playback unit can also extract specific patterns from the user's past playback history and customize the playback content. Furthermore, the playback unit can analyze the user's past playback history and prioritize playback of frequently played content. In this way, by customizing the playback content by referring to the user's past playback history, more appropriate audio can be played.

[0115] During playback, the playback unit can analyze the user's current environmental sound and play back at an optimal volume. The playback unit can, for example, analyze the environmental sound around the user and play back audio at an optimal volume. Furthermore, the playback unit can also play back audio at a lower volume when the user is in a quiet environment. Furthermore, the playback unit can also play back audio at a higher volume when the user is in a noisy environment. In this way, by analyzing the user's current environmental sound and playing back at an optimal volume, more appropriate audio can be played back.

[0116] The playback unit can estimate the user's emotion and determine the priority of audio to be played back based on the estimated user's emotion. For example, if the user is sad, the playback unit can estimate using the emotion engine and prioritize playing audio that requires emotional support. Also, if the user is feeling nostalgic, the playback unit can estimate using the emotion engine and prioritize playing audio related to memories. Furthermore, if the user is relaxed, the playback unit can estimate using the emotion engine and prioritize playing calm audio. In this way, by determining the priority of audio to be played back based on the user's emotion, more appropriate audio can be played back.

[0117] During playback, the playback unit can prioritize playback of highly relevant audio in consideration of the user's geographical location information. For example, if the user is in a specific area, the playback unit can prioritize playback of audio related to that area. Furthermore, if the user is traveling, the playback unit can prioritize playback of audio related to the travel destination. Furthermore, if the user is in the hometown of a deceased person, the playback unit can prioritize playback of audio related to that area. In this way, by prioritizing playback of highly relevant audio in consideration of the user's geographical location information, more appropriate audio can be played.

[0118] During playback, the playback unit can analyze the user's social media activity and play back related audio. For example, the playback unit can play back related audio based on content shared by the user on social media. The playback unit can also analyze the content posted by the user on social media and play back related audio. Furthermore, the playback unit can also play back related audio with reference to the activity of the user's friends on social media. In this way, by analyzing the user's social media activity and playing back related audio, more appropriate audio can be played back.

[0119] The playback unit can customize the playback method by reflecting the user's past feedback during playback. The playback unit can customize the playback method, for example, based on feedback provided by the user in the past. The playback unit can also adjust the playback means by reflecting the user's past feedback. Furthermore, the playback unit can customize the playback interface based on the user's past feedback. In this way, by customizing the playback method by reflecting the user's past feedback, more appropriate audio can be played back. === Hard Collateral 1-1 === Each of the multiple elements, including the collection unit, learning unit, reception unit, generation unit, playback unit, and emotion engine, is implemented, for example, by at least one of the smart device 14 and the data processing device 12. For example, the collection unit can collect video and audio data of the deceased using the camera 42 and microphone 38B of the smart device 14. The learning unit, implemented by the specific processing unit 290 of the data processing device 12, learns from the collected data and analyzes the deceased's voice and speaking style. The reception unit can accept user input using the touch panel 38A and microphone 38B of the smart device 14. The generation unit, implemented by the specific processing unit 290 of the data processing device 12, generates new words in the deceased's voice based on the user input. The playback unit can play back the generated words using the speaker 40B of the smart device 14. The emotion engine, implemented by the specific processing unit 290 of the data processing device 12, estimates the user's emotions and adjusts the type of data to be collected. === Hard Collateral 1-2 === Each of the multiple elements, including the collection unit, learning unit, reception unit, generation unit, playback unit, and emotion engine, described above, is realized, for example, by at least one of the smart glasses 214 and the data processing device 12. For example, the collection unit can collect video and audio data of the deceased using the camera 42 and microphone 238 of the smart glasses 214. The learning unit is realized by the specific processing unit 290 of the data processing device 12 and learns from the collected data and analyzes the deceased's voice and speaking style. The reception unit can accept user input using the microphone 238 of the smart glasses 214. The generation unit is realized by the specific processing unit 290 of the data processing device 12 and generates new words in the deceased's voice based on the user's input. The playback unit can play back the generated words using the speaker 240 of the smart glasses 214. The emotion engine is realized by the specific processing unit 290 of the data processing device 12 and estimates the user's emotions and adjusts the type of data to be collected. === Hard Collateral 1-3 === Each of the multiple elements, including the collection unit, learning unit, reception unit, generation unit, playback unit, and emotion engine, is realized, for example, by at least one of the headset-type terminal 314 and the data processing device 12. For example, the collection unit can collect video and audio data of the deceased using the camera 42 and microphone 238 of the headset-type terminal 314. The learning unit is realized by the specific processing unit 290 of the data processing device 12 and learns from the collected data and analyzes the deceased's voice and speaking style. The reception unit can receive user input using the microphone 238 of the headset-type terminal 314. The generation unit is realized by the specific processing unit 290 of the data processing device 12 and generates new words in the deceased's voice based on the user input. The playback unit can play back the generated words using the speaker 240 of the headset-type terminal 314. The emotion engine is realized by the specific processing unit 290 of the data processing device 12 and estimates the user's emotions and adjusts the type of data to be collected. === Hard Collateral 1-4 === Each of the multiple elements, including the collection unit, learning unit, reception unit, generation unit, playback unit, and emotion engine, is realized, for example, by at least one of the robot 414 and the data processing device 12. For example, the collection unit can collect video and audio data of the deceased using the camera 42 and microphone 238 of the robot 414. The learning unit is realized by the specific processing unit 290 of the data processing device 12 and learns from the collected data and analyzes the deceased's voice and speaking style. The reception unit can receive user input using the microphone 238 of the robot 414. The generation unit is realized by the specific processing unit 290 of the data processing device 12 and generates new words in the deceased's voice based on the user input. The playback unit can play back the generated words using the speaker 240 of the robot 414. The emotion engine is realized by the specific processing unit 290 of the data processing device 12 and estimates the user's emotions and adjusts the type of data to be collected.

[0120] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0121] The system further includes an emotion estimation unit that estimates the user's emotion and adjusts the content of the generated words based on the estimated emotion. For example, if the user is sad, the emotion estimation unit can instruct the generation unit to generate kind words or encouraging words. If the user is happy, the emotion estimation unit can instruct the generation unit to generate congratulatory words or words of sympathy. Furthermore, if the user is anxious, the emotion estimation unit can instruct the generation unit to generate reassuring words or encouraging words. In this way, appropriate words can be generated according to the user's emotion.

[0122] When collecting video and audio data of the deceased, the collection unit can diversify the sources of data collection. For example, the collection unit can collect video and audio data posted from the deceased's social media accounts. The collection unit can also collect data from archives of television and radio programs in which the deceased appeared. Furthermore, the collection unit can collect recordings of events and lectures in which the deceased participated. This allows for the collection of more diverse data and a wealth of material for recreating the deceased's voice and video.

[0123] The learning unit can also take into account the emotional expressions of the deceased when analyzing the characteristics of the deceased's voice and speaking patterns. For example, the learning unit can analyze the tone and rhythm of the deceased's voice when they are happy and build a model that generates new words based on that. It can also analyze the tone and rhythm of the deceased's voice when they are sad and build a model that generates new words based on that. It can also analyze the tone and rhythm of the deceased's voice when they are angry and build a model that generates new words based on that. This makes it possible to generate more natural-looking words that reflect the emotional expressions of the deceased.

[0124] The reception unit can adjust the tone and rhythm of the words generated in the voice of the deceased based on the content input by the user. For example, if the content input by the user is emotional, the reception unit can instruct the generation unit to generate words with a tone and rhythm that reflects the emotion. Also, if the content input by the user is a question, the reception unit can instruct the generation unit to generate a natural response. Furthermore, if the content input by the user is everyday conversation, the reception unit can instruct the generation unit to generate natural conversation. This makes it possible to generate appropriate words according to the content input by the user.

[0125] The generation unit can customize the content of the words generated in the voice of the deceased based on the user's input. For example, if the user speaks about a specific theme, the generation unit can generate words related to that theme. Also, if the user speaks about a specific event, the generation unit can generate words related to that event. Furthermore, if the user expresses a specific emotion, the generation unit can generate words corresponding to that emotion. In this way, customized words can be generated according to the user's input.

[0126] When reproducing the generated words, the reproduction unit can estimate the user's emotions and adjust the tone and rhythm of the reproduced audio based on the estimated emotions. For example, if the user is sad, the reproduction unit can reproduce audio with a gentle tone and a slow rhythm. If the user is happy, the reproduction unit can also reproduce audio with a bright tone and a rhythmic rhythm. Furthermore, if the user is relaxed, the reproduction unit can also reproduce audio with a gentle tone and a slow rhythm. This makes it possible to reproduce appropriate audio according to the user's emotions.

[0127] The system also includes a data analysis unit that can analyze the collected data and extract specific characteristics and patterns of the deceased. For example, the data analysis unit can extract phrases and expressions frequently used by the deceased. The data analysis unit can also analyze the words the deceased used in specific situations. Furthermore, the data analysis unit can analyze changes in the tone and rhythm of the deceased's voice and build a model to generate new words based on that analysis. This allows the system to generate more natural-sounding words that reflect the characteristics and patterns of the deceased.

[0128] The collection unit may diversify the data collection methods when collecting video and audio data of the deceased. For example, the collection unit may collect video and audio data provided by the deceased's family and friends. The collection unit may also collect data directly from devices used by the deceased. Furthermore, the collection unit may collect data from online communities and forums in which the deceased participated. This allows for the collection of more diverse data and enriches the material for recreating the deceased's voice and video.

[0129] The learning unit can also take into account the emotional expressions of the deceased when analyzing the characteristics of the deceased's voice and speaking patterns. For example, the learning unit can analyze the tone and rhythm of the deceased's voice when they are happy and build a model that generates new words based on that. It can also analyze the tone and rhythm of the deceased's voice when they are sad and build a model that generates new words based on that. It can also analyze the tone and rhythm of the deceased's voice when they are angry and build a model that generates new words based on that. This makes it possible to generate more natural-looking words that reflect the emotional expressions of the deceased.

[0130] When reproducing the generated words, the reproduction unit can estimate the user's emotions and adjust the tone and rhythm of the reproduced audio based on the estimated emotions. For example, if the user is sad, the reproduction unit can reproduce audio with a gentle tone and a slow rhythm. If the user is happy, the reproduction unit can also reproduce audio with a bright tone and a rhythmic rhythm. Furthermore, if the user is relaxed, the reproduction unit can also reproduce audio with a gentle tone and a slow rhythm. This makes it possible to reproduce appropriate audio according to the user's emotions.

[0131] The processing flow of the second embodiment will be briefly explained below.

[0132] Step 1: The collection unit collects video and audio data of the deceased. For example, the collection unit can collect video clips and audio messages of the deceased speaking. The collection unit can also collect photographs and text data of the deceased. Step 2: The learning unit learns from the data collected by the collection unit and analyzes the deceased's voice and speaking style. For example, the learning unit analyzes the deceased's voice characteristics and speaking patterns and builds a model to generate new words based on them. The learning unit can use generative AI (e.g., text generation AI or multimodal generation AI) to learn the deceased's voice characteristics and speaking patterns. Step 3: The reception unit receives input from the user. For example, the reception unit can input what the user will say to the deceased. The reception unit can support various input methods, such as voice input, text input, and image input. Step 4: The generation unit generates new words in the voice of the deceased based on the input received by the reception unit. For example, the generation unit can generate natural responses in the voice of the deceased based on the user's input. The generation unit can use a generation AI to generate new words in the voice of the deceased. Step 5: The playback unit plays back the words generated by the generation unit. For example, the playback unit can play back the generated response in the deceased's voice. The playback unit can select the optimal voice playback method taking into account the user's device information.

[0133] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0134] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AIs include the data generation model 58, such as a neural network model (e.g., a neural network model), and a neural network model (e.g., a neural network model). The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating speech, text data indicating text, and image data indicating an image is also input to the data generation model 58. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AIs other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0135] Furthermore, the processing by the data processing system 10 described above is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0136] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0137] [Second embodiment] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0138] 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0139] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0140] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0141] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0142] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0143] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0144] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0145] The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0146] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0147] In the smart glasses 214, the specific processing is performed by the processor 46. A specific processing program 60 is stored in the storage 50. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific processing program 60 executed on the RAM 48. The smart glasses 214 also have a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0148] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0149] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0150] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt containing an instruction, as well as inference data such as voice data representing speech, text data representing text, and image data representing an image. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AIs other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0151] The data processing system 210 according to the second embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 210 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the smart glasses 214 or an external device, etc., and the smart glasses 214 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0152] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0153] [Third embodiment] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0154] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0155] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0156] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0157] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0158] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0159] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0160] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0161] The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0162] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0163] In the headset type terminal 314, the identification process is performed by the processor 46. A identification program 60 is stored in the storage 50. The processor 46 reads the identification program 60 from the storage 50 and executes the read identification program 60 on the RAM 48. The identification process is realized by the processor 46 operating as a control unit 46A in accordance with the identification program 60 executed on the RAM 48. Note that the headset type terminal 314 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the identification processing unit 290 using these models.

[0164] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0165] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0166] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt containing an instruction, as well as inference data such as voice data representing speech, text data representing text, and image data representing an image. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AIs other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0167] The data processing system 310 according to the third embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 310 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the headset type terminal 314, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the headset type terminal 314. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the headset type terminal 314 or an external device, etc., and the headset type terminal 314 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0168] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0169] [Fourth embodiment] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0170] 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0171] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0172] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[0173] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0174] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS image sensor or a CCD image sensor, and captures images of the user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0175] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0176] The control object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[0177] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0178] The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0179] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0180] In the robot 414, the processor 46 performs the identification process. The storage 50 stores the identification program 60. The processor 46 reads the identification program 60 from the storage 50 and executes the read identification program 60 on the RAM 48. The identification process is realized by the processor 46 operating as the control unit 46A in accordance with the identification program 60 executed on the RAM 48. The robot 414 also has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can perform the same process as the identification processing unit 290 using these models.

[0181] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0182] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0183] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt containing an instruction, as well as inference data such as voice data representing speech, text data representing text, and image data representing an image. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AIs other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0184] The data processing system 410 according to the fourth embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 410 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the robot 414 or an external device, etc., and the robot 414 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0185] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0186] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0187] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion encompasses both emotions and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[0188] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[0189] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[0190] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is expressed, and when they approach the ideal, a state of pleasure is expressed. Emotions can also be created for robots, cars, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is expressed, and when they approach the ideal, a state of pleasure is expressed. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on speech emotion recognition and brain physiological signal analysis systems for emotions, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the area called "reaction," where sensation is dominant. The right half of the emotion map lists emotions belonging to the area called "situation," where situational awareness is dominant.

[0191] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[0192] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[0193] In the above embodiment, an example was given in which a specific process is performed by one computer 22, but the technology disclosed herein is not limited to this, and distributed processing of the specific process may be performed by multiple computers including computer 22.

[0194] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[0195] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[0196] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[0197] The hardware resource for executing a specific process can be any of the following types of processors: A processor, for example, is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. A processor also includes a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[0198] The hardware resource that executes the specific process may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific process may be a single processor.

[0199] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[0200] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[0201] In the above example, the first to fourth embodiments have been described separately, but some or all of these embodiments may be combined. The smart device 14, smart glasses 214, headset terminal 314, and robot 414 are merely examples, and they may be combined, or other devices may be used. In the above example, the first and second embodiments have been described separately, but they may be combined.

[0202] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[0203] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[0204] [Explanation of symbols]

[0205] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot

Claims

1. a collection unit that collects video and audio data of the deceased; a learning unit that learns the data collected by the collecting unit and analyzes the voice and speaking style of the deceased; a reception unit that receives input from a user; a generator that generates new words in the voice of the deceased based on the input received by the receiver; a reproduction unit that reproduces the words generated by the generation unit. A system characterized by:

2. The collecting unit Collect video clips or audio messages of the deceased speaking 2. The system of claim 1.

3. The learning unit Analyzing the voice characteristics or speaking patterns of the deceased and building a model to generate new words based on that 2. The system of claim 1.

4. The reception unit Users can input what they want to say to the deceased.

2. The system of claim 1.

5. The generation unit Generates appropriate responses in the deceased person's voice based on user input 2. The system of claim 1.

6. The playback unit Playing back generated responses 2. The system of claim 1.

7. The collecting unit It is possible to estimate the user's emotions and adjust the type of data to be collected based on the estimated user emotions.

2. The system of claim 1.

8. The collecting unit When collecting video and audio data of the deceased, assess the quality of the data and select appropriate data.

2. The system of claim 1.

Images