system

Abstract

The system according to this embodiment aims to convert speech into text and automatically correct grammatical and stylistic errors. [Solution] The system according to the embodiment comprises a conversion unit, a correction unit, and a learning unit. The conversion unit receives speech data during utterance and converts it into text data. The correction unit analyzes the text data generated by the conversion unit and corrects grammatical and stylistic errors. The learning unit accumulates and analyzes speech data to improve the accuracy of speech recognition.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, and includes steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the prior art, grammar and expression errors during voice-to-text conversion have not been fully automatically corrected, and there is room for improvement.

[0005] The system according to the embodiment aims to convert voice to text and automatically correct grammar and expression errors.

Means for Solving the Problems

[0006] The system according to this embodiment comprises a conversion unit, a correction unit, and a learning unit. The conversion unit receives speech data during utterance and converts it into text data. The correction unit analyzes the text data generated by the conversion unit and corrects grammatical and stylistic errors. The learning unit accumulates and analyzes speech data to improve the accuracy of speech recognition. [Effects of the Invention]

[0007] The system according to this embodiment can convert speech to text and automatically correct grammatical and stylistic errors. [Brief explanation of the drawing]

[0008] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Modes for carrying out the invention]

[0009] Hereinafter, an example of an embodiment of the system relating to the technology of this disclosure will be described with reference to the attached drawings.

[0010] First, let's explain the terminology used in the following explanation.

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

[0012] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

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

[0014] In the following embodiments, the labeled communication I / F (Interface) is an interface including a communication processor, an antenna, and the like. The communication I / F controls communication between a plurality of 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), or Bluetooth (registered trademark).

[0015] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B". That is, "A and / or B" means that it may be only A, only B, or a combination of A and B. Also, in this specification, when expressing three or more matters connected by "and / or", the same concept as "A and / or B" is applied.

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

[0017] As shown in FIG. 1, the 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, the RAM 30, and the storage 32 are connected to a bus 34. Also, the database 24 and the communication I / F 26 are 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 comprises a computer 36, a receiving device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The receiving device 38, output device 40, and camera 42 are also connected to the bus 52.

[0020] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, and accepts user input. The touch panel 38A accepts user input via touch by detecting contact with an object (e.g., a pen or finger). The microphone 38B accepts user input via voice by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 (see Figure 2) acquires the data indicating the user input.

[0021] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user by outputting the data in a form perceptible to the user (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0022] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

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

[0024] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0025] Storage 32 stores the data generation model 58 and the 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 the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0026] In the smart device 14, specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. The specific processing program 60 is used in conjunction 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 a control unit 46A according to the specific processing program 60 executed on the RAM 48. The smart device 14 also has a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0027] Furthermore, other devices besides the data processing device 12 may also 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 processing results (such as prediction results) using the data generation model 58 by communicating with the server device having the data generation model 58. The data processing device 12 may also be a server device or a terminal device owned by a user (e.g., a mobile phone, robot, home appliance, etc.). Next, an example of processing by the data processing system 10 according to the first embodiment will be described.

[0028] (Example of form 1) The AI ​​assistant for document creation support using speech recognition according to an embodiment of the present invention is a system that converts speech to text in real time and automatically corrects common grammatical and stylistic errors. This system receives speech data, converts it to text data, corrects grammatical or stylistic errors, and improves the accuracy of speech recognition. For example, by converting speech during a meeting to text in real time, meeting minutes can be automatically created. In this case, when converting the speech data, the system selects the optimal conversion method by referring to the user's past speech patterns. Furthermore, the conversion accuracy can be improved by adding a filtering function that automatically removes background noise. Next, the generated text data is analyzed and grammatical or stylistic errors are corrected. For example, by automatically detecting and correcting grammatical errors and unnatural expressions, more natural sentences can be generated. In this case, the system can learn the pronunciation or intonation of specific words to improve recognition accuracy in subsequent uses. Furthermore, the accuracy of speech recognition can be improved by accumulating and analyzing speech data. For example, by accumulating and analyzing the user's speech data, the accuracy of speech recognition in subsequent uses can be improved. It is also possible to estimate the user's emotions and adjust the accuracy of speech data conversion based on the estimated emotions of the user. This mechanism allows the system to learn the characteristics and quirks of the user's voice, improving accuracy. For example, by learning the pronunciation and intonation of a specific user, it becomes possible to perform speech recognition tailored to that user. Furthermore, real-time conversion of speech data and grammatical correction enable efficient document creation. This makes it usable for a variety of purposes, such as creating meeting minutes in situations like meetings and interviews, or taking everyday notes. As a result, the speech recognition-based AI assistant for document creation can convert speech data into text in real time, automatically correct grammatical and phrasing errors, and improve the accuracy of speech recognition.

[0029] The AI ​​assistant for document creation using speech recognition according to this embodiment comprises a conversion unit, a correction unit, and a learning unit. The conversion unit receives speech data during utterance and converts it into text data. The conversion unit can, for example, convert speech data into text data in real time. The conversion unit can also select the optimal conversion method by referring to the user's past speech patterns when converting speech data. The conversion unit can also have a filtering function that automatically removes background noise. For example, the conversion unit can automatically create meeting minutes by converting statements made during a meeting into text in real time. The conversion unit selects the optimal conversion method by referring to the user's past speech patterns. The conversion unit can have a filtering function that automatically removes background noise. The correction unit analyzes the text data generated by the conversion unit and corrects grammatical and stylistic errors. The correction unit can, for example, automatically detect and correct grammatical errors and unnatural expressions. The correction unit can also learn the pronunciation or intonation of specific words to improve recognition accuracy in subsequent uses. For example, the correction unit can automatically detect and correct grammatical errors. The correction unit learns the pronunciation or intonation of specific words to improve recognition accuracy in subsequent uses. The learning unit improves speech recognition accuracy by accumulating and analyzing audio data. For example, the learning unit can improve speech recognition accuracy in subsequent uses by accumulating and analyzing the user's audio data. The learning unit can also estimate the user's emotions and adjust the accuracy of audio data conversion based on the estimated emotions. For example, the learning unit accumulates and analyzes the user's audio data. The learning unit estimates the user's emotions and adjusts the accuracy of audio data conversion based on the estimated emotions. As a result, the speech recognition-based document creation support AI assistant according to this embodiment can convert audio data into text in real time, automatically correct grammatical and expression errors, and improve speech recognition accuracy.

[0030] The conversion unit receives speech data during utterance and converts it into text data. For example, the conversion unit can convert speech data into text data in real time. Specifically, the conversion unit uses an advanced speech recognition algorithm to analyze the speech signal and break it down into phonemes and words. This enables rapid and accurate conversion of speech data into text data. The conversion unit can also select the optimal conversion method by referring to the user's past speech patterns. For example, if the user frequently uses specific technical terms or proper nouns, the unit learns these patterns to improve conversion accuracy in subsequent uses. Furthermore, the conversion unit can also include a filtering function that automatically removes background noise. This allows for high-accuracy speech recognition and text conversion even in noisy environments. For example, by converting meeting statements into text in real time, meeting minutes can be automatically created. The conversion unit selects the optimal conversion method by referring to the user's past speech patterns. This enables optimal conversion tailored to the user's speech style and habits, improving conversion accuracy. Furthermore, the conversion unit includes a filtering function that automatically removes background noise. This allows for high-accuracy speech recognition and text conversion even in noisy environments. For example, meeting minutes can be automatically created by converting spoken words in a meeting into text in real time. The conversion unit selects the optimal conversion method by referring to the user's past speech patterns. This enables optimal conversion tailored to the user's speech style and habits, improving conversion accuracy. Furthermore, the conversion unit includes a filtering function that automatically removes background noise. This allows for highly accurate speech recognition and text conversion even in noisy environments. For example, meeting minutes can be automatically created by converting spoken words in a meeting into text in real time.

[0031] The correction unit analyzes the text data generated by the conversion unit and corrects grammatical and stylistic errors. For example, the correction unit can automatically detect and correct grammatical errors and unnatural expressions. Specifically, it uses natural language processing techniques to analyze the grammatical structure of the text data and identify grammatical errors. For example, it checks subject-verb agreement, tense agreement, and appropriate punctuation. The correction unit can also learn the pronunciation or intonation of specific words to improve recognition accuracy in subsequent uses. For example, if a user pronounces a particular word with a unique pronunciation, the correction unit learns that pronunciation pattern and incorporates it into subsequent speech recognition. This allows the correction unit to improve the quality of the text data generated by the conversion unit. Furthermore, the correction unit can also perform context-aware corrections. For example, it can correct homonym errors based on context. This allows the meaning of the text data to be conveyed more accurately. This allows for optimal corrections tailored to the user's pronunciation and intonation habits, improving correction accuracy. Furthermore, the correction unit can also perform context-aware corrections. For example, it can correct homonym errors based on context. This ensures that the meaning of the text data is conveyed more accurately. The correction unit learns the pronunciation or intonation of specific words, improving recognition accuracy for subsequent uses. This allows for optimal corrections tailored to the user's pronunciation and intonation habits, improving correction accuracy. Furthermore, the correction unit can also perform context-aware corrections. For example, it can correct homonym errors based on context. This ensures that the meaning of the text data is conveyed more accurately.

[0032] The learning unit improves the accuracy of speech recognition by accumulating and analyzing audio data. For example, by accumulating and analyzing the user's audio data, the learning unit can improve the accuracy of speech recognition in subsequent uses. Specifically, the learning unit uses deep learning technology to extract features from the audio data and continuously updates the speech recognition model. This allows it to learn the user's pronunciation and intonation habits and reflect them in subsequent speech recognition. Furthermore, the learning unit can estimate the user's emotions and adjust the accuracy of the audio data conversion based on the estimated emotions. For example, it adjusts the speech recognition parameters according to changes in emotions, such as when the user is tense or relaxed. This enables the learning unit to achieve optimal speech recognition tailored to the user's emotional state. The learning unit accumulates and analyzes the user's voice data. This allows it to learn the user's pronunciation and intonation habits, which can then be reflected in subsequent voice recognition. Furthermore, the learning unit estimates the user's emotions and adjusts the accuracy of voice data conversion based on the estimated emotions. This enables optimal voice recognition tailored to the user's emotional state. The learning unit accumulates and analyzes the user's voice data. This allows it to learn the user's pronunciation and intonation habits, which can then be reflected in subsequent voice recognition. Furthermore, the learning unit estimates the user's emotions and adjusts the accuracy of voice data conversion based on the estimated emotions. This enables optimal voice recognition tailored to the user's emotional state.

[0033] The conversion unit can convert audio data into text data in real time. For example, by converting speech during a meeting into text in real time, the conversion unit can automatically create meeting minutes. The conversion unit can also select the optimal conversion method by referring to the user's past speech patterns. The conversion unit can also include a filtering function that automatically removes background noise. This allows for immediate document creation by converting audio data into text in real time.

[0034] The learning unit can improve the accuracy of speech recognition in subsequent uses by accumulating and analyzing the user's voice data. For example, the learning unit can improve the accuracy of speech recognition in subsequent uses by accumulating and analyzing the user's voice data. The learning unit can also estimate the user's emotions and adjust the accuracy of voice data conversion based on the estimated emotions. This allows for improved speech recognition accuracy through the accumulation and analysis of the user's voice data.

[0035] The correction unit can learn the pronunciation and intonation of specific words to improve recognition accuracy in subsequent uses. For example, the correction unit can learn the pronunciation and intonation of specific words to improve recognition accuracy in subsequent uses. The correction unit can also automatically detect and correct grammatical errors and unnatural expressions. This improves recognition accuracy in subsequent uses by learning the pronunciation and intonation of specific words.

[0036] The conversion unit can select the optimal conversion method by referring to the user's past speech patterns when converting audio data. For example, the conversion unit's AI learns phrases and expressions that the user has frequently used in the past and improves conversion accuracy based on that. The conversion unit analyzes the user's past speech patterns and adjusts the conversion method considering specific pronunciation habits. If the user frequently uses certain technical terms, the conversion unit selects a conversion method that prioritizes the recognition of those terms. In this way, conversion accuracy is improved by referring to the user's past speech patterns.

[0037] The conversion unit can be equipped with a filtering function that automatically removes background noise during audio data conversion. For example, the conversion unit's AI can automatically detect and remove background noise in a conference room. The conversion unit's AI can analyze speech in noisy environments and filter out the noise to improve conversion accuracy. When a user speaks while moving, the conversion unit's AI can remove wind noise and traffic noise before converting the audio data. This automatically removes background noise, improving conversion accuracy.

[0038] The conversion unit can prioritize the recognition of region-specific expressions by considering the user's geographical location information when converting voice data. For example, if the user uses a dialect from a specific region, the AI ​​will prioritize the recognition of that dialect. If the user is traveling, the AI ​​will prioritize the recognition of local place names and specific expressions. If the user has a cultural background from a specific region, the AI ​​will recognize expressions specific to that region. This improves conversion accuracy by prioritizing the recognition of region-specific expressions.

[0039] The conversion unit analyzes the user's social media activity during audio data conversion and prioritizes the recognition of relevant terms. For example, the conversion unit's AI learns the social media terms the user frequently uses and prioritizes their recognition during conversion. If the user frequently uses a particular hashtag, the conversion unit's AI prioritizes the recognition of that hashtag. Based on the user's social media activity, the conversion unit's AI recognizes specific trending terms. As a result, analyzing social media activity improves the accuracy of relevant term recognition.

[0040] The editing function can match writing style and expressions by referring to the user's past document creation history when correcting grammar. For example, the editing function can learn the writing style of documents the user has created in the past and match it when correcting grammar. If the user frequently uses certain expressions, the editing function will prioritize using those expressions when correcting grammar. If the user prefers a certain writing style, the editing function will match that style when correcting grammar. In this way, matching writing style and expressions becomes possible by referring to the user's past document creation history.

[0041] The correction function can prioritize correcting region-specific expressions by considering the user's geographical location during grammatical correction. For example, if the user uses a dialect from a specific region, the AI ​​will prioritize correcting that dialect. If the user is traveling, the AI ​​will prioritize correcting local place names and specific expressions. If the user has a cultural background from a specific region, the AI ​​will correct expressions specific to that region. This improves the accuracy of grammatical correction by prioritizing the correction of region-specific expressions.

[0042] The correction unit can analyze the user's social media activity during grammatical correction and prioritize correcting relevant terminology. For example, the correction unit's AI learns the social media terms the user frequently uses and prioritizes their use during grammatical correction. If the user frequently uses a particular hashtag, the correction unit's AI will prioritize using that hashtag during grammatical correction. Based on the user's social media activity, the correction unit's AI will use specific trending terms during grammatical correction. This improves the accuracy of relevant terminology correction by analyzing social media activity.

[0043] The learning unit can optimize its learning algorithm by referring to past learning data during the learning process. For example, the learning unit can optimize the AI's learning algorithm based on past learning data to improve recognition accuracy. The learning unit can optimize the AI's learning algorithm by referring to audio data previously spoken by the user. If the user frequently uses certain words, the learning unit can optimize the AI's learning algorithm based on the audio data of those words. In this way, the learning algorithm can be optimized by referring to past learning data.

[0044] The learning unit can prioritize learning region-specific expressions by considering the user's geographical location during the learning process. For example, if the user uses a dialect of a particular region, the AI ​​will prioritize learning that dialect. If the user is traveling, the AI ​​will prioritize learning local place names and unique expressions. If the user has a cultural background of a particular region, the AI ​​will learn expressions specific to that region. This improves the accuracy of learning by prioritizing the learning of region-specific expressions.

[0045] The learning unit can analyze the user's social media activity during training and prioritize learning relevant terminology. For example, the learning unit can learn social media terms frequently used by the user, improving recognition accuracy in subsequent uses. If the user frequently uses a particular hashtag, the learning unit can learn that hashtag, improving recognition accuracy in subsequent uses. Based on the user's social media activity, the learning unit can learn specific trending terms, improving recognition accuracy in subsequent uses. In this way, analyzing social media activity improves the accuracy of learning relevant terminology.

[0046] The system according to the embodiment is not limited to the example described above, and various modifications are possible, for example, as follows.

[0047] The AI ​​assistant for document creation using speech recognition can also be equipped with a data analysis unit. This unit analyzes the text data generated by the conversion unit to analyze the user's speech patterns and trends. For example, it can convert speech during a meeting into text in real time, extract and analyze the user's speech patterns from that text. The data analysis unit can also select the optimal analysis method by referring to the user's past speech data. For example, it can analyze the frequency of use of specific keywords or phrases to improve the accuracy of future analyses. This allows for more effective document creation support by understanding the user's speech patterns and trends.

[0048] The AI ​​assistant for document creation using voice recognition can also be equipped with a schedule management unit. This unit analyzes the text data generated by the conversion unit and automatically manages the user's schedule. For example, it can convert meeting speeches into text in real time, extract important dates and tasks from the text, and add them to the schedule. The schedule management unit can also select the optimal management method by referring to the user's past schedule history. For example, it can prioritize adding specific tasks or events to the schedule, improving the accuracy of management in the future. This streamlines the user's schedule management.

[0049] The AI ​​assistant for document creation using voice recognition can also be equipped with a content generation unit. This unit automatically generates new content based on the text data generated by the conversion unit. For example, it can convert speech during a meeting into text in real time and automatically generate reports and presentation materials based on that text. The content generation unit can also select the optimal generation method by referring to the user's past content generation history. For example, it can prioritize the use of specific formats and styles to improve generation accuracy in subsequent uses. This streamlines the user's content creation process.

[0050] The AI ​​assistant for document creation using voice recognition can also be equipped with a data backup unit. This unit periodically backs up the text data generated by the conversion unit, ensuring data security. For example, it can convert meeting speeches into text in real time and automatically back up that text to cloud storage. The data backup unit can also select the optimal backup method by referring to the user's past backup history. For example, it can perform backups at specific times or frequencies to improve the accuracy of subsequent backups. This ensures the secure protection of the user's data.

[0051] The following briefly describes the processing flow for example form 1.

[0052] Step 1: The conversion unit receives the spoken audio data and converts it into text data. The conversion unit can convert audio data into text data in real time and can also select the optimal conversion method by referring to the user's past speech patterns. It can also add a filtering function to automatically remove background noise. For example, it can convert speech during a meeting into text in real time and automatically create meeting minutes. Step 2: The correction unit analyzes the text data generated by the conversion unit and corrects grammatical and stylistic errors. The correction unit can automatically detect and correct grammatical errors and unnatural expressions. It can also learn the pronunciation or intonation of specific words to improve recognition accuracy in subsequent uses. Step 3: The learning unit improves the accuracy of speech recognition by accumulating and analyzing voice data. The learning unit can improve the accuracy of speech recognition in subsequent uses by accumulating and analyzing the user's voice data. It can also estimate the user's emotions and adjust the accuracy of voice data conversion based on the estimated emotions of the user.

[0053] (Example of form 2) The AI ​​assistant for document creation support using speech recognition according to an embodiment of the present invention is a system that converts speech to text in real time and automatically corrects common grammatical and stylistic errors. This system receives speech data, converts it to text data, corrects grammatical or stylistic errors, and improves the accuracy of speech recognition. For example, by converting speech during a meeting to text in real time, meeting minutes can be automatically created. In this case, when converting the speech data, the system selects the optimal conversion method by referring to the user's past speech patterns. Furthermore, the conversion accuracy can be improved by adding a filtering function that automatically removes background noise. Next, the generated text data is analyzed and grammatical or stylistic errors are corrected. For example, by automatically detecting and correcting grammatical errors and unnatural expressions, more natural sentences can be generated. In this case, the system can learn the pronunciation or intonation of specific words to improve recognition accuracy in subsequent uses. Furthermore, the accuracy of speech recognition can be improved by accumulating and analyzing speech data. For example, by accumulating and analyzing the user's speech data, the accuracy of speech recognition in subsequent uses can be improved. It is also possible to estimate the user's emotions and adjust the accuracy of speech data conversion based on the estimated emotions of the user. This mechanism allows the system to learn the characteristics and quirks of the user's voice, improving accuracy. For example, by learning the pronunciation and intonation of a specific user, it becomes possible to perform speech recognition tailored to that user. Furthermore, real-time conversion of speech data and grammatical correction enable efficient document creation. This makes it usable for a variety of purposes, such as creating meeting minutes in situations like meetings and interviews, or taking everyday notes. As a result, the speech recognition-based AI assistant for document creation can convert speech data into text in real time, automatically correct grammatical and phrasing errors, and improve the accuracy of speech recognition.

[0054] The AI ​​assistant for document creation using speech recognition according to this embodiment comprises a conversion unit, a correction unit, and a learning unit. The conversion unit receives speech data during utterance and converts it into text data. The conversion unit can, for example, convert speech data into text data in real time. The conversion unit can also select the optimal conversion method by referring to the user's past speech patterns when converting speech data. The conversion unit can also have a filtering function that automatically removes background noise. For example, the conversion unit can automatically create meeting minutes by converting statements made during a meeting into text in real time. The conversion unit selects the optimal conversion method by referring to the user's past speech patterns. The conversion unit can have a filtering function that automatically removes background noise. The correction unit analyzes the text data generated by the conversion unit and corrects grammatical and stylistic errors. The correction unit can, for example, automatically detect and correct grammatical errors and unnatural expressions. The correction unit can also learn the pronunciation or intonation of specific words to improve recognition accuracy in subsequent uses. For example, the correction unit can automatically detect and correct grammatical errors. The correction unit learns the pronunciation or intonation of specific words to improve recognition accuracy in subsequent uses. The learning unit improves speech recognition accuracy by accumulating and analyzing audio data. For example, the learning unit can improve speech recognition accuracy in subsequent uses by accumulating and analyzing the user's audio data. The learning unit can also estimate the user's emotions and adjust the accuracy of audio data conversion based on the estimated emotions. For example, the learning unit accumulates and analyzes the user's audio data. The learning unit estimates the user's emotions and adjusts the accuracy of audio data conversion based on the estimated emotions. As a result, the speech recognition-based document creation support AI assistant according to this embodiment can convert audio data into text in real time, automatically correct grammatical and expression errors, and improve speech recognition accuracy.

[0055] The conversion unit receives speech data during utterance and converts it into text data. For example, the conversion unit can convert speech data into text data in real time. Specifically, the conversion unit uses an advanced speech recognition algorithm to analyze the speech signal and break it down into phonemes and words. This enables rapid and accurate conversion of speech data into text data. The conversion unit can also select the optimal conversion method by referring to the user's past speech patterns. For example, if the user frequently uses specific technical terms or proper nouns, the unit learns these patterns to improve conversion accuracy in subsequent uses. Furthermore, the conversion unit can also include a filtering function that automatically removes background noise. This allows for high-accuracy speech recognition and text conversion even in noisy environments. For example, by converting meeting statements into text in real time, meeting minutes can be automatically created. The conversion unit selects the optimal conversion method by referring to the user's past speech patterns. This enables optimal conversion tailored to the user's speech style and habits, improving conversion accuracy. Furthermore, the conversion unit includes a filtering function that automatically removes background noise. This allows for high-accuracy speech recognition and text conversion even in noisy environments. For example, meeting minutes can be automatically created by converting spoken words in a meeting into text in real time. The conversion unit selects the optimal conversion method by referring to the user's past speech patterns. This enables optimal conversion tailored to the user's speech style and habits, improving conversion accuracy. Furthermore, the conversion unit includes a filtering function that automatically removes background noise. This allows for highly accurate speech recognition and text conversion even in noisy environments. For example, meeting minutes can be automatically created by converting spoken words in a meeting into text in real time.

[0056] The correction unit analyzes the text data generated by the conversion unit and corrects grammatical and stylistic errors. For example, the correction unit can automatically detect and correct grammatical errors and unnatural expressions. Specifically, it uses natural language processing techniques to analyze the grammatical structure of the text data and identify grammatical errors. For example, it checks subject-verb agreement, tense agreement, and appropriate punctuation. The correction unit can also learn the pronunciation or intonation of specific words to improve recognition accuracy in subsequent uses. For example, if a user pronounces a particular word with a unique pronunciation, the correction unit learns that pronunciation pattern and incorporates it into subsequent speech recognition. This allows the correction unit to improve the quality of the text data generated by the conversion unit. Furthermore, the correction unit can also perform context-aware corrections. For example, it can correct homonym errors based on context. This allows the meaning of the text data to be conveyed more accurately. This allows for optimal corrections tailored to the user's pronunciation and intonation habits, improving correction accuracy. Furthermore, the correction unit can also perform context-aware corrections. For example, it can correct homonym errors based on context. This ensures that the meaning of the text data is conveyed more accurately. The correction unit learns the pronunciation or intonation of specific words, improving recognition accuracy for subsequent uses. This allows for optimal corrections tailored to the user's pronunciation and intonation habits, improving correction accuracy. Furthermore, the correction unit can also perform context-aware corrections. For example, it can correct homonym errors based on context. This ensures that the meaning of the text data is conveyed more accurately.

[0057] The learning unit improves the accuracy of speech recognition by accumulating and analyzing audio data. For example, by accumulating and analyzing the user's audio data, the learning unit can improve the accuracy of speech recognition in subsequent uses. Specifically, the learning unit uses deep learning technology to extract features from the audio data and continuously updates the speech recognition model. This allows it to learn the user's pronunciation and intonation habits and reflect them in subsequent speech recognition. Furthermore, the learning unit can estimate the user's emotions and adjust the accuracy of the audio data conversion based on the estimated emotions. For example, it adjusts the speech recognition parameters according to changes in emotions, such as when the user is tense or relaxed. This enables the learning unit to achieve optimal speech recognition tailored to the user's emotional state. The learning unit accumulates and analyzes the user's voice data. This allows it to learn the user's pronunciation and intonation habits, which can then be reflected in subsequent voice recognition. Furthermore, the learning unit estimates the user's emotions and adjusts the accuracy of voice data conversion based on the estimated emotions. This enables optimal voice recognition tailored to the user's emotional state. The learning unit accumulates and analyzes the user's voice data. This allows it to learn the user's pronunciation and intonation habits, which can then be reflected in subsequent voice recognition. Furthermore, the learning unit estimates the user's emotions and adjusts the accuracy of voice data conversion based on the estimated emotions. This enables optimal voice recognition tailored to the user's emotional state.

[0058] The conversion unit can convert audio data into text data in real time. For example, by converting speech during a meeting into text in real time, the conversion unit can automatically create meeting minutes. The conversion unit can also select the optimal conversion method by referring to the user's past speech patterns. The conversion unit can also include a filtering function that automatically removes background noise. This allows for immediate document creation by converting audio data into text in real time.

[0059] The learning unit can improve the accuracy of speech recognition in subsequent uses by accumulating and analyzing the user's voice data. For example, the learning unit can improve the accuracy of speech recognition in subsequent uses by accumulating and analyzing the user's voice data. The learning unit can also estimate the user's emotions and adjust the accuracy of voice data conversion based on the estimated emotions. This allows for improved speech recognition accuracy through the accumulation and analysis of the user's voice data.

[0060] The correction unit can learn the pronunciation and intonation of specific words to improve recognition accuracy in subsequent uses. For example, the correction unit can learn the pronunciation and intonation of specific words to improve recognition accuracy in subsequent uses. The correction unit can also automatically detect and correct grammatical errors and unnatural expressions. This improves recognition accuracy in subsequent uses by learning the pronunciation and intonation of specific words.

[0061] The conversion unit can estimate the user's emotions and adjust the accuracy of the audio data conversion based on the estimated emotions. For example, if the user is nervous, the conversion unit will more precisely analyze subtle differences in pronunciation to improve the accuracy of the audio data conversion. If the user is relaxed, the conversion unit will adjust the accuracy of the audio data conversion, prioritizing natural pronunciation. If the user is in a hurry, the conversion unit will adjust the accuracy of the audio data conversion, using a simplified conversion method to quickly convert to text. This allows for more accurate text conversion by adjusting the accuracy of the audio data conversion based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI includes, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI.

[0062] The conversion unit can select the optimal conversion method by referring to the user's past speech patterns when converting audio data. For example, the conversion unit's AI learns phrases and expressions that the user has frequently used in the past and improves conversion accuracy based on that. The conversion unit analyzes the user's past speech patterns and adjusts the conversion method considering specific pronunciation habits. If the user frequently uses certain technical terms, the conversion unit selects a conversion method that prioritizes the recognition of those terms. In this way, conversion accuracy is improved by referring to the user's past speech patterns.

[0063] The conversion unit can be equipped with a filtering function that automatically removes background noise during audio data conversion. For example, the conversion unit's AI can automatically detect and remove background noise in a conference room. The conversion unit's AI can analyze speech in noisy environments and filter out the noise to improve conversion accuracy. When a user speaks while moving, the conversion unit's AI can remove wind noise and traffic noise before converting the audio data. This automatically removes background noise, improving conversion accuracy.

[0064] The conversion unit can estimate the user's emotions and adjust the expression of the converted text based on the estimated emotions. For example, if the user is angry, the AI ​​will soften the expression of the text. If the user is happy, the AI ​​will add positive expressions to the text. If the user is sad, the AI ​​will soften the expression of the text. This allows for more appropriate expression by adjusting the expression of the text based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI is a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to these examples.

[0065] The conversion unit can prioritize the recognition of region-specific expressions by considering the user's geographical location information when converting voice data. For example, if the user uses a dialect from a specific region, the AI ​​will prioritize the recognition of that dialect. If the user is traveling, the AI ​​will prioritize the recognition of local place names and specific expressions. If the user has a cultural background from a specific region, the AI ​​will recognize expressions specific to that region. This improves conversion accuracy by prioritizing the recognition of region-specific expressions.

[0066] The conversion unit analyzes the user's social media activity during audio data conversion and prioritizes the recognition of relevant terms. For example, the conversion unit's AI learns the social media terms the user frequently uses and prioritizes their recognition during conversion. If the user frequently uses a particular hashtag, the conversion unit's AI prioritizes the recognition of that hashtag. Based on the user's social media activity, the conversion unit's AI recognizes specific trending terms. As a result, analyzing social media activity improves the accuracy of relevant term recognition.

[0067] The editing unit can estimate the user's emotions and correct grammatical or stylistic errors based on those emotions. For example, if the user is nervous, the AI ​​will strictly correct grammatical errors. If the user is relaxed, the AI ​​will prioritize naturalness of expression in its corrections. If the user is in a hurry, the AI ​​will quickly correct grammatical errors. This results in more natural-sounding text by correcting grammatical and stylistic errors based on the user's emotions. Emotion estimation is achieved using emotion estimation functions, such as emotion engines or generative AI. Generative AI includes, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI.

[0068] The editing function can match writing style and expressions by referring to the user's past document creation history when correcting grammar. For example, the editing function can learn the writing style of documents the user has created in the past and match it when correcting grammar. If the user frequently uses certain expressions, the editing function will prioritize using those expressions when correcting grammar. If the user prefers a certain writing style, the editing function will match that style when correcting grammar. In this way, matching writing style and expressions becomes possible by referring to the user's past document creation history.

[0069] The correction function can prioritize correcting region-specific expressions by considering the user's geographical location during grammatical correction. For example, if the user uses a dialect from a specific region, the AI ​​will prioritize correcting that dialect. If the user is traveling, the AI ​​will prioritize correcting local place names and specific expressions. If the user has a cultural background from a specific region, the AI ​​will correct expressions specific to that region. This improves the accuracy of grammatical correction by prioritizing the correction of region-specific expressions.

[0070] The correction unit can analyze the user's social media activity during grammatical correction and prioritize correcting relevant terminology. For example, the correction unit's AI learns the social media terms the user frequently uses and prioritizes their use during grammatical correction. If the user frequently uses a particular hashtag, the correction unit's AI will prioritize using that hashtag during grammatical correction. Based on the user's social media activity, the correction unit's AI will use specific trending terms during grammatical correction. This improves the accuracy of relevant terminology correction by analyzing social media activity.

[0071] The learning unit can estimate the user's emotions and select training data based on the estimated emotions. For example, if the user is nervous, the learning unit will prioritize learning audio data from relaxed states. If the user is relaxed, the learning unit will prioritize learning audio data with natural pronunciation. If the user is in a hurry, the learning unit will prioritize learning audio data with rapid pronunciation. This improves the accuracy of learning by selecting training data based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AIs include, but are not limited to, text generation AI (e.g., LLM) or multimodal generation AI.

[0072] The learning unit can optimize its learning algorithm by referring to past learning data during the learning process. For example, the learning unit can optimize the AI's learning algorithm based on past learning data to improve recognition accuracy. The learning unit can optimize the AI's learning algorithm by referring to audio data previously spoken by the user. If the user frequently uses certain words, the learning unit can optimize the AI's learning algorithm based on the audio data of those words. In this way, the learning algorithm can be optimized by referring to past learning data.

[0073] The learning unit can prioritize learning region-specific expressions by considering the user's geographical location during the learning process. For example, if the user uses a dialect of a particular region, the AI ​​will prioritize learning that dialect. If the user is traveling, the AI ​​will prioritize learning local place names and unique expressions. If the user has a cultural background of a particular region, the AI ​​will learn expressions specific to that region. This improves the accuracy of learning by prioritizing the learning of region-specific expressions.

[0074] The learning unit can analyze the user's social media activity during training and prioritize learning relevant terminology. For example, the learning unit can learn social media terms frequently used by the user, improving recognition accuracy in subsequent uses. If the user frequently uses a particular hashtag, the learning unit can learn that hashtag, improving recognition accuracy in subsequent uses. Based on the user's social media activity, the learning unit can learn specific trending terms, improving recognition accuracy in subsequent uses. In this way, analyzing social media activity improves the accuracy of learning relevant terminology.

[0075] The system according to the embodiment is not limited to the example described above, and various modifications are possible, for example, as follows.

[0076] AI assistants that use speech recognition to assist with document creation can also be equipped with a translation unit. The translation unit translates the text data generated by the conversion unit into other languages. For example, by converting speech during a meeting into text in real time and instantly translating that text into other languages, it can automatically create multilingual meeting minutes. The translation unit can also select the optimal translation method by referring to the user's past translation history. For example, it can learn the translation of specific technical terms and phrases to improve translation accuracy in the future. Furthermore, the translation unit can estimate the user's emotions and adjust the tone and nuance of the translation based on the estimated emotions. For example, if the user is angry, the translation unit can soften the tone of the text. This improves the accuracy and appropriateness of the translation based on the user's emotions.

[0077] The AI ​​assistant for document creation using speech recognition can also be equipped with a summarization function. The summarization function analyzes the text data generated by the conversion function, extracts key points, and creates a summary. For example, it can convert speech during a meeting into text in real time and summarize that text to concisely summarize the main topics and conclusions of the meeting. The summarization function can also select the optimal summarization method by referring to the user's past summarization history. For example, it can prioritize the extraction of specific keywords or phrases to improve the accuracy of future summaries. Furthermore, the summarization function can estimate the user's emotions and adjust the tone and content of the summary based on the estimated emotions. For example, if the user is relaxed, the summarization function can soften the expression in the text. This improves the accuracy and appropriateness of the summary based on the user's emotions.

[0078] The AI ​​assistant for document creation using voice recognition can also be equipped with a notification unit. The notification unit analyzes the text data generated by the conversion unit, extracts important information and action items, and notifies the user. For example, it can convert speech during a meeting into text in real time, extract important tasks and deadlines from that text, and remind the user. The notification unit can also refer to the user's past notification history to select the optimal notification method. For example, it can prioritize extracting specific keywords or phrases to improve the accuracy of future notifications. Furthermore, the notification unit can estimate the user's emotions and adjust the timing and content of notifications based on those emotions. For example, if the user is in a hurry, the notification unit can quickly notify them of important information. This improves the accuracy and appropriateness of notifications based on the user's emotions.

[0079] The AI ​​assistant for document creation using speech recognition can also be equipped with a feedback unit. The feedback unit analyzes the text data generated by the conversion unit and provides real-time feedback to the user. For example, it can convert speech during a meeting into text in real time and suggest improvements to grammar and expression in that text. The feedback unit can also select the optimal feedback method by referring to the user's past feedback history. For example, it can prioritize pointing out specific grammatical errors or unnatural expressions to improve the accuracy of feedback in the future. Furthermore, the feedback unit can estimate the user's emotions and adjust the tone and content of the feedback based on those emotions. For example, if the user is nervous, the feedback unit can provide feedback in a calmer tone. This improves the accuracy and appropriateness of the feedback based on the user's emotions.

[0080] The AI ​​assistant for document creation using voice recognition can also be equipped with a suggestion function. This function analyzes the text data generated by the conversion function and provides improvement suggestions to the user. For example, it can convert meeting speeches into text in real time and suggest more effective expressions and structures for that text. The suggestion function can also select the optimal suggestion method by referring to the user's past suggestion history. For example, it can prioritize suggesting improvements to specific expressions or structures to improve the accuracy of future suggestions. Furthermore, the suggestion function can estimate the user's emotions and adjust the tone and content of suggestions based on those emotions. For example, if the user is relaxed, the suggestion function can provide suggestions in a softer tone. This improves the accuracy and appropriateness of suggestions based on the user's emotions.

[0081] The AI ​​assistant for document creation using voice recognition can also include a customization feature. This customization feature adjusts system settings based on the user's individual needs and preferences. For example, it can be set to prioritize the recognition of specific technical terms or phrases. The customization feature can also refer to the user's past setting history to select the optimal settings. For example, it can be set to prioritize the use of specific writing styles or expressions to improve recognition accuracy in subsequent uses. Furthermore, the customization feature can estimate the user's emotions and adjust settings based on those emotions. For example, if the user is relaxed, the customization feature can provide flexible settings. This optimizes the system settings based on the user's emotions.

[0082] The AI ​​assistant for document creation using speech recognition can also be equipped with a data analysis unit. This unit analyzes the text data generated by the conversion unit to analyze the user's speech patterns and trends. For example, it can convert speech during a meeting into text in real time, extract and analyze the user's speech patterns from that text. The data analysis unit can also select the optimal analysis method by referring to the user's past speech data. For example, it can analyze the frequency of use of specific keywords or phrases to improve the accuracy of future analyses. This allows for more effective document creation support by understanding the user's speech patterns and trends.

[0083] The AI ​​assistant for document creation using voice recognition can also be equipped with a schedule management unit. This unit analyzes the text data generated by the conversion unit and automatically manages the user's schedule. For example, it can convert meeting speeches into text in real time, extract important dates and tasks from the text, and add them to the schedule. The schedule management unit can also select the optimal management method by referring to the user's past schedule history. For example, it can prioritize adding specific tasks or events to the schedule, improving the accuracy of management in the future. This streamlines the user's schedule management.

[0084] The AI ​​assistant for document creation using voice recognition can also be equipped with a content generation unit. This unit automatically generates new content based on the text data generated by the conversion unit. For example, it can convert speech during a meeting into text in real time and automatically generate reports and presentation materials based on that text. The content generation unit can also select the optimal generation method by referring to the user's past content generation history. For example, it can prioritize the use of specific formats and styles to improve generation accuracy in subsequent uses. This streamlines the user's content creation process.

[0085] The AI ​​assistant for document creation using voice recognition can also be equipped with a data backup unit. This unit periodically backs up the text data generated by the conversion unit, ensuring data security. For example, it can convert meeting speeches into text in real time and automatically back up that text to cloud storage. The data backup unit can also select the optimal backup method by referring to the user's past backup history. For example, it can perform backups at specific times or frequencies to improve the accuracy of subsequent backups. This ensures the secure protection of the user's data.

[0086] The following briefly describes the processing flow for example form 2.

[0087] Step 1: The conversion unit receives the spoken audio data and converts it into text data. The conversion unit can convert audio data into text data in real time and can also select the optimal conversion method by referring to the user's past speech patterns. It can also add a filtering function to automatically remove background noise. For example, it can convert speech during a meeting into text in real time and automatically create meeting minutes. Step 2: The correction unit analyzes the text data generated by the conversion unit and corrects grammatical and stylistic errors. The correction unit can automatically detect and correct grammatical errors and unnatural expressions. It can also learn the pronunciation or intonation of specific words to improve recognition accuracy in subsequent uses. Step 3: The learning unit improves the accuracy of speech recognition by accumulating and analyzing voice data. The learning unit can improve the accuracy of speech recognition in subsequent uses by accumulating and analyzing the user's voice data. It can also estimate the user's emotions and adjust the accuracy of voice data conversion based on the estimated emotions of the user.

[0088] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0089] Data generation model 58 is a form of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AI include text generation AI, image generation AI, and multimodal generation AI. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats from audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVMs), k-means clustering, convolutional neural networks (CNNs), recurrent neural networks (RNNs), generative adversarial networks (GANs), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each of the above parts is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example.Furthermore, processing performed by AI, including generative AI, may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by AI, including generative AI.

[0090] Furthermore, the processing performed by the data processing system 10 described above is carried out by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but it may also be carried out by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. In addition, 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.

[0091] For example, the conversion unit is implemented by the processor 46 of the smart device 14, and converts speech data into text data in real time. The correction unit is implemented by the specific processing unit 290 of the data processing device 12, and analyzes the generated text data to correct grammatical and expression errors. The learning unit is implemented by the specific processing unit 290 of the data processing device 12, and improves the accuracy of speech recognition by accumulating and analyzing speech data. The conversion unit, correction unit, and learning unit can also be implemented by the control unit 46A of the smart device 14, for example. The correspondence between each unit and the device or control unit is not limited to the example described above, and various changes are possible.

[0092] [Second Embodiment] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0093] As shown in Figure 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.

[0094] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN and / or LAN.

[0095] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0096] The microphone 238 receives voice signals from the user and accepts instructions from the user. The microphone 238 captures the voice signals from the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0097] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, which captures images of the area around the user (for example, an imaging range defined by a field of view equivalent to the field of vision of a typical healthy person).

[0098] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0099] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing by the processor 28. The storage 32 stores the specific processing program 56.

[0100] The processor 28 reads a 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 acting as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0101] Storage 32 stores the data generation model 58 and the 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 the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0102] In the smart glasses 214, specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. 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 acting as a control unit 46A according to the specific processing program 60 executed on the RAM 48. The smart glasses 214 also have a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0103] Furthermore, other devices besides the data processing device 12 may also 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 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device that has the data generation model 58. Also, the data processing device 12 may be a server device or a terminal device owned by the user (for example, a mobile phone, robot, home appliance, etc.).

[0104] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0105] The data generation model 58 is a so-called generative AI. An example of a 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 is input with prompts containing instructions, and inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats such as audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each part described above is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example. Also, processing performed by an AI including a generative AI may be replaced by rule-based processing, and rule-based processing may be replaced by processing performed by an AI including a generative AI.

[0106] 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 performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but it may also be performed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart glasses 214 or an external device, and the smart glasses 214 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0107] For example, the conversion unit is implemented by the processor 46 of the smart glasses 214, which converts voice data into text data in real time. The correction unit is implemented by the specific processing unit 290 of the data processing device 12, which analyzes the generated text data and corrects grammatical and expression errors. The learning unit is implemented by the specific processing unit 290 of the data processing device 12, which improves the accuracy of voice recognition by accumulating and analyzing voice data. The conversion unit, correction unit, and learning unit can also be implemented by the control unit 46A of the smart glasses 214, for example. The correspondence between each unit and the device or control unit is not limited to the examples described above, and various modifications are possible.

[0108] [Third Embodiment] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0109] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0110] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN and / or LAN.

[0111] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0112] The microphone 238 receives voice signals from the user and accepts instructions from the user. The microphone 238 captures the voice signals from the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0113] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, which captures images of the area around the user (for example, an imaging range defined by a field of view equivalent to the field of vision of a typical healthy person).

[0114] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0115] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0116] The processor 28 reads a 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 acting as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0117] Storage 32 stores the data generation model 58 and the 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 the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0118] In the headset terminal 314, specific processing is performed by the processor 46. The storage 50 stores a specific program 60. The processor 46 reads the specific program 60 from the storage 50 and executes the read specific program 60 on the RAM 48. The specific processing is realized by the processor 46 acting as a control unit 46A according to the specific program 60 executed on the RAM 48. The headset terminal 314 also has a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0119] Furthermore, other devices besides the data processing device 12 may also 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 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device that has the data generation model 58. Also, the data processing device 12 may be a server device or a terminal device owned by the user (for example, a mobile phone, robot, home appliance, etc.).

[0120] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0121] The data generation model 58 is a so-called generative AI. An example of a 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 is input with prompts containing instructions, and inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats such as audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each part described above is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example. Also, processing performed by an AI including a generative AI may be replaced by rule-based processing, and rule-based processing may be replaced by processing performed by an AI including a generative AI.

[0122] 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 performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the headset terminal 314, but may also be performed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the headset terminal 314. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the headset terminal 314 or an external device, and the headset terminal 314 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0123] For example, the conversion unit is implemented by the processor 46 of the headset terminal 314, which converts voice data into text data in real time. The correction unit is implemented by the specific processing unit 290 of the data processing device 12, which analyzes the generated text data and corrects grammatical and expression errors. The learning unit is implemented by the specific processing unit 290 of the data processing device 12, which improves the accuracy of voice recognition by accumulating and analyzing voice data. The conversion unit, correction unit, and learning unit can also be implemented by the control unit 46A of the headset terminal 314, for example. The correspondence between each unit and the device or control unit is not limited to the example described above, and various changes are possible.

[0124] [Fourth Embodiment] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0125] As shown in Figure 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.

[0126] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN and / or LAN.

[0127] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0128] The microphone 238 receives voice signals from the user and accepts instructions from the user. The microphone 238 captures the voice signals from the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0129] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS image sensor or CCD image sensor, which captures images of the area around the user (for example, an imaging range defined by a field of view equivalent to the field of vision of a typical healthy person).

[0130] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0131] The controlled 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 robot 414's emotions can be expressed by controlling these motors. The robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0132] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0133] The processor 28 reads a 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 acting as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0134] Storage 32 stores the data generation model 58 and the 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 the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0135] In robot 414, specific processing is performed by processor 46. A specific program 60 is stored in storage 50. Processor 46 reads the specific program 60 from storage 50 and executes it on RAM 48. The specific processing is achieved by processor 46 acting as a control unit 46A according to the specific program 60 executed on RAM 48. Robot 414 also has data generation model 58 and emotion identification model 59, similar to those of the robot, and can perform processing similar to that of the specific processing unit 290 using these models.

[0136] Furthermore, other devices besides the data processing device 12 may also 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 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device that has the data generation model 58. Also, the data processing device 12 may be a server device or a terminal device owned by the user (for example, a mobile phone, robot, home appliance, etc.).

[0137] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0138] The data generation model 58 is a so-called generative AI. An example of a 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 is input with prompts containing instructions, and inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats such as audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each part described above is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example. Also, processing performed by an AI including a generative AI may be replaced by rule-based processing, and rule-based processing may be replaced by processing performed by an AI including a generative AI.

[0139] 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 performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but it may also be performed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the robot 414 or an external device, and the robot 414 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0140] For example, the conversion unit is implemented by the processor 46 of the robot 414, which converts speech data into text data in real time. The correction unit is implemented by the specific processing unit 290 of the data processing device 12, which analyzes the generated text data and corrects grammatical and expression errors. The learning unit is implemented by the specific processing unit 290 of the data processing device 12, which improves the accuracy of speech recognition by accumulating and analyzing speech data. The conversion unit, correction unit, and learning unit can also be implemented by the control unit 46A of the robot 414, for example. The correspondence between each unit and the device or control unit is not limited to the example described above, and various modifications are possible.

[0141] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0142] Figure 9 shows the emotion map 400, in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0143] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0144] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0145] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, and motorcycles, emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated based, for example, on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0146] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0147] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0148] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing method for the specific process may be used, which includes computer 22 and multiple other computers.

[0149] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0150] 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.

[0151] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0152] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0153] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0154] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0155] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0156] Furthermore, although the above-described examples were divided into four embodiments, some or all of these embodiments may be combined. Also, the smart device 14, smart glasses 214, headset terminal 314, and robot 414 are just examples, and they may be combined, or other devices may be used. Also, although the above-described examples were divided into two embodiments, Embodiment 1 and Embodiment 2, these may be combined.

[0157] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and other things that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0158] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0159] (Note 1) A conversion unit that receives speech data during utterance and converts it into text data, A correction unit analyzes the text data generated by the conversion unit and corrects grammatical and stylistic errors, The system includes a learning unit for improving the accuracy of speech recognition by accumulating and analyzing the aforementioned audio data. A system characterized by the following features. (Note 2) The conversion unit is Convert audio data to text data in real time. The system described in Appendix 1, characterized by the features described herein. (Note 3) The aforementioned learning unit, By accumulating and analyzing the user's voice data, the accuracy of voice recognition in subsequent uses can be improved. The system described in Appendix 1, characterized by the features described herein. (Note 4) The aforementioned modification section is, Learn the pronunciation and intonation of specific words to improve recognition accuracy in subsequent uses. The system described in Appendix 1, characterized by the features described herein. (Note 5) The conversion unit is The system estimates the user's emotions and adjusts the accuracy of the voice data conversion based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 6) The conversion unit is When converting audio data, the system selects the optimal conversion method by referring to the user's past speech patterns. The system described in Appendix 1, characterized by the features described herein. (Note 7) The conversion unit is Add a filtering function that automatically removes background noise when converting audio data. The system described in Appendix 1, characterized by the features described herein. (Note 8) The conversion unit is It estimates the user's emotions and adjusts the way the converted text is expressed based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 9) The conversion unit is When converting audio data, region-specific expressions are prioritized based on the user's geographical location information. The system described in Appendix 1, characterized by the features described herein. (Note 10) The conversion unit is During the conversion of audio data, the system analyzes the user's social media activity and prioritizes the recognition of relevant terms. The system described in Appendix 1, characterized by the features described herein. (Note 11) The aforementioned modification section is, It estimates the user's emotions and corrects grammatical or stylistic errors based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 12) The aforementioned modification section is, When correcting grammar, the system refers to the user's past document creation history to match the writing style and expression. The system described in Appendix 1, characterized by the features described herein. (Note 13) The aforementioned modification section is, When correcting grammar, prioritize correcting region-specific expressions by taking into account the user's geographical location. The system described in Appendix 1, characterized by the features described herein. (Note 14) The aforementioned modification section is, When correcting grammar, the system analyzes the user's social media activity and prioritizes correcting relevant terminology. The system described in Appendix 1, characterized by the features described herein. (Note 15) The aforementioned learning unit, The system estimates the user's emotions and selects training data based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 16) The aforementioned learning unit, During training, the learning algorithm is optimized by referring to past training data. The system described in Appendix 1, characterized by the features described herein. (Note 17) The aforementioned learning unit, During learning, the system prioritizes learning region-specific expressions, taking into account the user's geographical location. The system described in Appendix 1, characterized by the features described herein. (Note 18) The aforementioned learning unit, During learning, the system analyzes the user's social media activity and prioritizes learning relevant terminology. The system described in Appendix 1, characterized by the features described herein. [Explanation of Symbols]

[0160] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots

Claims

1. A conversion unit that receives speech data during utterance and converts it into text data, A correction unit analyzes the text data generated by the conversion unit and corrects grammatical and stylistic errors, The system includes a learning unit for improving the accuracy of speech recognition by accumulating and analyzing the aforementioned audio data. A system characterized by the following features.

2. The conversion unit is Convert audio data to text data in real time. The system according to feature 1.

3. The aforementioned learning unit, By accumulating and analyzing the user's voice data, the accuracy of voice recognition in subsequent uses can be improved. The system according to feature 1.

4. The aforementioned modification section is, Learn the pronunciation and intonation of specific words to improve recognition accuracy in subsequent uses. The system according to feature 1.

5. The conversion unit is The system estimates the user's emotions and adjusts the accuracy of the voice data conversion based on those estimated emotions. The system according to feature 1.

6. The conversion unit is When converting audio data, the system selects the optimal conversion method by referring to the user's past speech patterns. The system according to feature 1.

7. The conversion unit is Add a filtering function that automatically removes background noise when converting audio data. The system according to feature 1.

8. The conversion unit is It estimates the user's emotions and adjusts the way the converted text is expressed based on those estimated emotions. The system according to feature 1.

Images