Real-time automated modern Japanese translation distribution system and distribution method for audio recordings of Buddhist sutra chanting.

Abstract

The meaning of the chanting during a funeral service is conveyed to attendees in real time. [Solution] A distribution system for delivering modern Japanese translations of Buddhist chanting to attendees' information terminals at a funeral, comprising an audio acquisition unit, an audio recognition processing unit, a natural language processing engine, and a distribution server. The audio acquisition unit sequentially acquires the audio of the chanting at the funeral. The audio recognition processing unit recognizes the audio of the chanting and outputs text data. The natural language processing engine translates the text data into modern Japanese. The distribution server has a control unit and sequentially distributes the modern Japanese translations derived from the chanting to the attendees' information terminals.

Description

[Technical Field]

[0001] The present disclosure relates to a real-time automatic modern-language translation delivery system and delivery method for sutra-chanting voice. [Background Art]

[0002] In a Buddhist funeral, sutras chanted by monks are recited in ancient Japanese, Chinese, and / or Sanskrit. Therefore, the ceremony proceeds formally without most of the modern attendees being able to understand the content. For this reason, it is difficult for the teachings of Buddhism and the meaning of prayers to be conveyed to attendees, and funerals tend to become formal ceremonies. Accordingly, there has been proposed an apparatus that, in a funeral hall, pairs the Chinese translation of sutras being chanted with an explanation thereof and displays the pair on a large display device or a smartphone (see, for example, Patent Document 1). [Prior Art Literature] [Patent Literature]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2025-097426 [Summary of Invention] [Problem to be Solved by Invention]

[0004] However, in the prior art, the length of the explanation has been adjusted in advance so that the time is approximately the same as the sutra-chanting time for each sutra, instead of corresponding to the actual sutra-chanting time during the funeral. That is, in the prior art, there is no mechanism for synchronizing the sutra chanted by the monk with the content displayed on the display device, and the content displayed is also prepared in advance. Therefore, the content does not necessarily match the actual content of the sutra-chanting voice chanted by the monk. Accordingly, the prior art has room for improvement in terms of enabling attendees to understand the meaning of the monk's sutra chanting in real time and to receive the spirit of the prayer.

[0005] Accordingly, an object of the present disclosure, which has been made focusing on these points, is to provide a delivery system and a delivery method that convey the meaning of the monk's sutra-chanting voice to attendees in real time. [Means for solving the problem]

[0006] The invention of a distribution system according to one embodiment of the present disclosure is A distribution system that delivers a modern Japanese translation of Buddhist scriptures to the information terminals of attendees at a funeral, The aforementioned funeral includes an audio acquisition unit that sequentially acquires the audio of the chanting of sutras, A speech recognition processing unit that recognizes the audio of the chanting and outputs text data, A natural language processing engine that translates the aforementioned text data into the aforementioned modern Japanese translation. A distribution server having a control unit that sequentially distributes the modern Japanese translation obtained from the chanting of sutras to the information terminals of the attendees. It is equipped with.

[0007] The distribution method according to one embodiment of this disclosure is: A distribution method implemented by a distribution system that delivers a modern Japanese translation of Buddhist scriptures to the information terminals of attendees at a funeral, The audio of the chanting of sutras during the aforementioned funeral will be acquired sequentially, The system recognizes the audio of the chanting and outputs text data. Translating the aforementioned text data into the aforementioned modern Japanese translation, This includes sequentially distributing the modern Japanese translation derived from the chanting to the information terminals of the attendees. [Effects of the Invention]

[0008] According to this disclosure, the control unit sequentially delivers a modern Japanese translation of the chanting audio to the attendees' information terminals, thereby allowing the meaning of the chanting audio by the monks to be conveyed to the attendees in real time. [Brief explanation of the drawing]

[0009] [Figure 1] This figure shows a funeral venue and equipment configuration using a distribution system according to one embodiment. [Figure 2] This is a block diagram showing the schematic configuration of the distribution system according to the first embodiment. [Figure 3] Figure 2 illustrates the process when additional learning is performed on the speech recognition processing unit. [Figure 4] Figure 2 is a flowchart of the distribution process performed by the distribution system. [Figure 5] Figure 4 is a flowchart showing the procedure for "Generating a Modern Japanese Translation of Buddhist Chanting." [Figure 6] This figure shows an example of the display on the attendee's terminal. [Figure 7] This is a block diagram showing the schematic configuration of the distribution system according to the second embodiment. [Figure 8] This flowchart shows the procedure for "generating a modern Japanese translation of Buddhist scriptures" in the second embodiment. [Modes for carrying out the invention]

[0010] The embodiments of this disclosure will be described below with reference to the drawings. In the following, "scripture" refers to a text that compiles the teachings of Buddhism. Chanting means reading the text of the scripture aloud. In this application, the sound produced as a result of chanting is referred to as the sound of chanting. Chanting during a funeral may include individual elements such as the deceased's posthumous Buddhist name and family name. In addition, the content of the chanting may not always be fixed in advance, due to factors such as the repetition of scriptures according to the time allotted for offering incense.

[0011] (System Overview) Figure 1 shows a simplified top view of a funeral hall 50 using the distribution system 1 according to the embodiment of this disclosure, and an example of the equipment configuration. In the example in Figure 1, an altar 51 is provided in the front center of the funeral hall 50, and in front of it, starting from the altar 51 side, are a monk's seat 52 for chanting sutras and an incense burning stand 53. Behind the incense burning stand 53 are several seats for bereaved family members 54, and behind those are several seats for attendees 55. Also, a reception desk 56 is located outside the entrance of the funeral hall 50. The arrangement inside the funeral hall 50 is not limited to this, and various arrangements are possible. The capacity of the funeral hall 50 can also be set arbitrarily.

[0012] A microphone 21 is arranged near the monk's seat 52 at a position where the monk's sutra-chanting voice can be collected. The microphone 21 functions as a voice acquisition unit. In addition, an operation unit 22 for operating the distribution server 10 described later may be arranged inside or outside the funeral venue 50. The microphone 21 and the operation unit 22 are connected via a communication device 23 to the distribution system 1 located within the building where the funeral venue 50 is located or at a remote location. The communication device 23 may be a computer such as a PC (Personal Computer) having a communication function.

[0013] Funeral attendees carry information terminals such as smartphones. The information terminal carried by a funeral attendee is referred to as an attendee terminal 30 hereinafter. A funeral attendee can read code information such as a QR code 31 installed at the reception 56, the bereaved family seat 54, or the attendee seat 55. The information of the QR code 31 includes a URL (Uniform Resource Locator) of the distribution system 1 and information for identifying the funeral that the funeral attendee attends. The information for identifying the funeral may be information indicating the date, time and location, and may also be information represented by numbers. The QR code 31 may be attached to tables and / or seats and / or distributed as printed matter. When a funeral attendee reads the QR code 31 with the attendee terminal 30, the attendee terminal 30 is connected to the distribution server 10 (described later) of the distribution system 1 via a network 40 such as the Internet, and an application such as a web application is activated. This enables the distribution system 1 to distribute modern Japanese translations of sutras to the attendee terminal 30. The funeral attendee can view the attendee terminal 30 on their lap or the like during the funeral.

[0014] Next, an overview of the operation of the distribution system 1 will be described. When a monk starts sutra chanting at the funeral venue 50, the audio is sequentially transmitted to the distribution system 1 via the microphone 21 and the communication device 23. The distribution system 1 converts the audio of the monk's sutra chanting into text data, and further translates this text data into modern Japanese. The distribution system 1 sequentially distributes the modern Japanese translation of the sutra chanting audio to the attendee terminals 30 via the network 40. The distributed modern Japanese translation of the sutra chanting is sequentially displayed on the display unit of the attendee terminal 30. This enables funeral attendees to check the modern Japanese translation of the sutra chanting displayed on the attendee terminal 30 in real time while listening to the monk's sutra chanting audio.

[0015] The present distribution system 1 provides the modern Japanese translation of the sutra chanting audio to funeral attendees, so that funeral attendees can understand the meaning and content of the sutra chanting audio in real time. This allows funeral attendees to send off the deceased while inwardly accepting the deceased's life and the teachings of Buddhism. In this way, the distribution system 1 of the present disclosure can provide funeral attendees with a new funeral experience different from conventional ones.

[0016] (First Embodiment) Hereinafter, a more detailed configuration of the distribution system 1 according to the first embodiment will be described based on Fig. 2. The distribution system 1 includes an audio acquisition unit such as a microphone 21, a speech recognition processing unit 15 (STT: Speech-to-Text), a natural language processing engine 16, and a distribution server 10. The distribution system 1 may further include a speech synthesis processing unit 17. Note that in the distribution system 1, instead of the microphone 21, a component such as a receiving device that acquires remote audio signals may be regarded as the audio acquisition unit.

[0017] The speech recognition processing unit 15 converts the digital signal of the monk's chanting, acquired from the microphone 21, into text data. Hereinafter, the data obtained by converting the chanting audio into text will be referred to as "chanting text." Buddhist scriptures were originally written in Sanskrit (Brahma) and other languages ​​in ancient India, translated into Chinese in China, and then transmitted to Japan. Therefore, the chanting text output by the speech recognition processing unit 15 is primarily Chinese text found in Buddhist scriptures.

[0018] As the speech recognition processing unit 15, an AI speech recognition system utilizing artificial intelligence (AI) can be used. The AI ​​speech recognition system may be an on-premise system or a cloud-based system. A general-purpose AI speech recognition system that is commonly used can output a reasonably accurate Chinese scripture from the input chanting audio if the AI's training data includes Buddhist terminology and scriptures.

[0019] However, general-purpose AI speech recognition systems suffer from a deterioration in recognition accuracy due to factors such as the unique intonation, melody, speech rate variations, high number of homonyms, and unclear word boundaries inherent in Buddhist chanting. To address this problem, the speech recognition processing unit 15 can improve the recognition accuracy of Buddhist chanting by using a speech recognition model that has undergone additional training on a general-purpose AI speech recognition model. Here, additional training refers to fine-tuning. General-purpose AI speech recognition models include, but are not limited to, OpenAI Whisper and Qwen3-ASR.

[0020] The procedure for additional training of the general-purpose speech recognition model is explained with reference to Figure 3. First, the general-purpose speech recognition model is prepared (step S101). Next, many pairs of audio recordings of actual Buddhist monks chanting and corresponding correct texts are prepared to serve as data for additional training (step S102). Finally, the general-purpose speech recognition model is further trained using the data prepared in step S102 (step S103). By using the speech recognition model that has undergone this additional training, the recognition accuracy of Buddhist chanting by the speech recognition processing unit 15 can be improved.

[0021] The natural language processing engine 16 translates the chanting text output by the speech recognition processing unit 15 into a modern Japanese translation. The natural language processing engine 16 may be a base model trained on a large amount of text data using deep learning. The base model may be either on-premise or cloud-based. The base model may include large language models (LLMs) and vision language models (VLMs) that can process images and videos in addition to text. Alternatively, the natural language processing engine 16 may be a small language model (SLM) that is lighter than a large language model (LLM) and is specialized for the task of translating chanting texts.

[0022] The natural language processing engine 16 is given a prompt in advance to translate the input sutra text into modern Japanese. The prompt may include instructions to perform a correct translation based on the surrounding text data if it is determined that the sutra text contains a misrecognized character. This makes it possible to correct misrecognitions by the speech recognition processing unit 15 even at the stage of generating the modern Japanese translation by the natural language processing engine 16. The natural language processing engine 16 may use an AI model obtained by performing additional training on the base model to generate a modern Japanese translation from the sutra text.

[0023] The distribution server 10 is a computer for server use. For example, the distribution server 10 is a server belonging to a cloud computing system or other computing system. However, the distribution server 10 is not limited to these and may be any general-purpose electronic device such as a PC (Personal Computer), or other electronic device dedicated to the distribution system 1. Furthermore, the distribution server 10 can be installed at the funeral venue rather than in a remote location.

[0024] The distribution server 10 includes a communication unit 11, a control unit 12, and a storage unit 13.

[0025] The communication unit 11 includes at least one external communication interface for communicating with the communication device 23 and connecting to the network 40. The communication interface may be either a wired or wireless communication interface. In the case of wired communication, the communication interface may be, for example, a LAN (Local Area Network) interface or a USB (Universal Serial Bus) interface. In the case of wireless communication, the communication interface may be, for example, an interface compatible with mobile communication standards such as 4G (4th generation), 5G (5th generation), and 6G (6th generation).

[0026] The control unit 12 includes at least one processor, at least one dedicated circuit, or a combination thereof. The processor is a general-purpose processor such as a CPU (central processing unit) or GPU (graphics processing unit), or a dedicated processor specialized for a specific process. The dedicated circuit is, for example, an FPGA (field-programmable gate array) or an ASIC (application-specific integrated circuit). The control unit 12 controls each part of the distribution server 10 and executes processes related to the operation of the distribution server 10.

[0027] For example, when the control unit 12 receives an access request from a funeral attendee's terminal 30 that has read the QR code 31 via the communication unit 11, it launches a web application for distributing funeral information. The control unit 12 also distributes a modern Japanese translation of the chanting audio received from the natural language processing engine 16 to the attendee's terminal 30 that is accessing the distribution server 10.

[0028] The storage unit 13 includes semiconductor memory, magnetic memory, optical memory, or a combination of at least two of these. The storage unit 13 functions, for example, as main memory, auxiliary memory, or cache memory. The storage unit 13 stores programs and data used for the operation of the distribution server 10, and data obtained by the operation of the distribution server 10.

[0029] The memory unit 13 stores a funeral calendar, including the funeral schedule. The distribution server 10 manages the distribution of multiple funerals as separate distribution channels according to the funeral calendar. The memory unit 13 may also store the distribution URL and access permission information for each funeral on a per-distribution channel basis.

[0030] Although the distribution server 10 in Figure 2 of this embodiment does not show an input unit and an output unit, it may further include an input unit and an output unit. That is, in addition to receiving (input) and transmitting (output) information via the communication unit 11, the distribution server 10 may also perform information input and output using the input unit and output unit. The input unit may be, for example, a keyboard, mouse, and touch panel. The output unit may be, for example, a display such as an LCD (liquid crystal display) or an organic EL (electro-luminescence).

[0031] In Figure 2, the distribution server 10 is shown connected in series with the speech recognition processing unit 15 and the natural language processing engine 16. However, the distribution server 10 may also be positioned as a hub that performs information processing for the distribution system 1. That is, the distribution server 10 may acquire chanting audio data from the microphone 21, transmit the acquired audio data to the speech recognition processing unit 15, and receive the chanting text from the speech recognition processing unit 15. The distribution server 10 may also transmit the chanting text to the natural language processing engine 16 and receive a modern Japanese translation from the natural language processing engine 16.

[0032] The functions of the distribution server 10 are realized by executing a program relating to the information processing method of this embodiment on a processor corresponding to the control unit 12. In other words, the functions of the distribution server 10 are realized by software. The program causes the computer to perform the operations of the distribution server 10, thereby causing the computer to function as the distribution server 10. That is, the computer functions as the distribution server 10 by performing the operations of the distribution server 10 according to the program.

[0033] In this embodiment, the program can be recorded on a computer-readable recording medium. The computer-readable recording medium includes non-temporary computer-readable media, such as magnetic recording devices, optical discs, magneto-optical recording media, or semiconductor memory. The program can be distributed, for example, by selling, transferring, or lending portable recording media such as DVDs (digital versatile discs) or CD-ROMs (compact disc read-only memory) on which the program is recorded. Alternatively, the program may be distributed by storing it on the storage of an external server and transmitting it from the external server to other computers. The program may also be provided as a program product.

[0034] The speech synthesis processing unit 17 is a system that generates speech from text. The speech synthesis processing unit 17 may be a system that uses AI speech synthesis technology. The speech synthesis processing unit 17 obtains a modern Japanese translation of the chanting audio generated by the natural language processing engine 16 from the distribution server 10 and synthesizes it into speech. The speech synthesis processing unit 17 transmits the synthesized speech to the attendee terminal 30 of a specific funeral attendee via the distribution server 10. For example, attendees may choose to receive a modern Japanese translation of the chanting audio from the attendee terminal 30. In this way, it becomes possible to provide a modern Japanese translation of the chanting audio to attendees who have difficulty reading the display on the attendee terminal 30 due to impaired vision, etc.

[0035] Next, an example of the process performed by the distribution system 1 will be explained based on the flowchart in Figure 4.

[0036] First, attendees arriving at the funeral venue 50 scan a QR code 31 placed at the reception area 56, the bereaved family seating area 54, or the attendee seating area 55 using their attendee terminal 30. This causes the attendee terminal 30 to access the URL of the distribution server 10 indicated by the QR code 31 (step S201).

[0037] The control unit 12 of the distribution server 10 compares the information identifying the funeral information contained in the QR code 31 with the funeral calendar in the storage unit 13, and connects the access from the attendee terminal 30 to the appropriate funeral distribution channel (step S202).

[0038] The control unit 12 transmits information about the dedicated web page for the appropriate distribution channel to the attendee terminal 30, causing it to be displayed on the display unit of the attendee terminal 30 (step S203). Steps S201 to S203 complete the preparations before the start of the funeral. Funeral attendees do not need to install a dedicated app or perform manual input to view the dedicated page.

[0039] Next, the funeral service begins, and the monk starts chanting. Prior to the start of the funeral service, the distribution system 1 may be set to a state where it can acquire audio by operating the control unit 22. This operation of the control unit 22 is not mandatory. When the monk starts chanting, the microphone 21 converts the chanting audio into an electrical signal and transmits it to the distribution system 1. As a result, the distribution system 1 acquires the audio of the monk chanting (step S204).

[0040] In the distribution system 1, the speech recognition processing unit 15 sequentially converts the acquired chanting audio into text (step S205). The chanting text generated at this time will be the Chinese text used in the scriptures. For example, when the audio input to the speech recognition processing unit 15 is the sound "Kanjizai Bosatsu", the text "Kanjizai Bosatsu" is output. However, due to the unique way monks pronounce words and sounds other than chanting, such as the wooden fish drum, the speech recognition result may contain some misrecognition. The method for reducing the rate of misrecognition by the speech recognition processing unit 15 is as described above. The speech recognition processing unit 15 outputs the chanting text to the natural language processing engine 16.

[0041] The natural language processing engine 16 generates a modern Japanese text translation of the chanting audio (step S206). The process in step S206 is explained below with reference to Figure 5. First, the natural language processing engine 16 is given instructions in advance to translate the sequentially input chanting text into modern Japanese. The instructions also include instructions to correct any errors in the chanting text and output a modern Japanese translation if it is determined that there are any misrecognitions.

[0042] In this state, the speech recognition processing unit 15 inputs the chanting text to the natural language processing engine 16 (step S301).

[0043] The natural language processing engine 16 sequentially translates the input chanting text into modern Japanese (step S302). Based on the instructions in the instruction sentence, the natural language processing engine 16 can correct any misrecognitions in the chanting text and generate a modern Japanese translation. The natural language processing engine 16 may also adjust the length of the generated modern Japanese translation according to the length of the chanting text. For example, if the modern Japanese translation is longer than the chanting text, it may take attendees too long to read the translation and they may not be able to keep up with the progress of the chanting audio. Therefore, the natural language processing engine 16 may summarize and output the modern Japanese translation according to the speed of the chanting. The natural language processing engine 16 may also generate the modern Japanese translation in short sentences to make it easier for attendees to understand.

[0044] The natural language processing engine 16 sequentially outputs the generated modern Japanese translation to the distribution server 10 (step S303).

[0045] Returning to the explanation of the flowchart in Figure 4, the control unit 12 of the distribution server 10 assigns display settings to the attendee terminal 30 for the modern Japanese translation of the chanting audio generated in step S206 (step S207). For example, the display settings align the latest modern Japanese translation to a predetermined position on the attendee terminal 30, and as a result scroll the previously distributed modern Japanese translations upward. This ensures that the latest modern Japanese translation of the chanting audio is always displayed in an easily viewable position on the attendee terminal 30, and substantially suppresses scrolling of the display screen by attendees. The display settings may also include an instruction to highlight only the latest distributed modern Japanese translation. In addition to scrolling and highlighting the distributed modern Japanese translations, the control unit 12 can specify colors and / or use bold text.

[0046] The control unit 12 sequentially transmits the modern Japanese translations, with the display settings assigned, to the attendee terminals 30 (step S208). The attendee terminals 30 sequentially display the modern Japanese translations of the transmitted chanting audio at predetermined positions on the display unit, according to the display settings. This eliminates the need for attendees to move their eyes significantly to the attendee terminals 30 during the funeral service. The control unit 12 controls the web application on the attendee terminals 30 to maintain silence without emitting any sound.

[0047] The process from steps S204 to S208 is repeated while the monk is chanting (step S209: No). This allows funeral attendees to see the modern Japanese translation of the scripture currently being chanted in real time.

[0048] An example of the display on the attendee terminal 30 is shown in Figure 6. In this example, the name of the scripture being chanted is displayed at the top of the attendee terminal 30. The name of the scripture being chanted may be recognized by the natural language processing engine 16 and sent to the distribution server 10. The control unit 12 of the distribution server 10 may instruct the attendee terminal 30 to set the display to show the name of the scripture at a fixed position at the top of the attendee terminal 30's display and send this instruction to the attendee terminal 30. The modern Japanese translation of the chanting may be scrolled sequentially, and the latest distributed content may be fixedly displayed in a clearly visible predetermined position in the center or bottom of the display.

[0049] Furthermore, the control unit 12 of the distribution server 10 may change the screen displayed on the display unit of the attendee terminal 30 according to the progress of the funeral. For example, the control unit 12 may display a graphic on the attendee terminal 30 indicating the elapsed time since the start of chanting. Also, based on the output from the natural language processing engine 16, the control unit 12 may display text on a part of the display unit of the attendee terminal 30 during events such as incense burning and / or silent prayer.

[0050] As described above, according to this embodiment, a modern Japanese translation of the chanting can be displayed in real time on the attendee terminal 30 in conjunction with the monk's chanting. This allows attendees to understand the meaning of the monk's chanting in real time and receive the spirit of prayer. Furthermore, the speech recognition processing unit 15 and / or natural language processing engine correct any misrecognitions of the chanting voice, thereby reducing the occurrence of recognition errors. Moreover, since the modern Japanese translation is displayed in an easily viewable position and manner on the display unit of the attendee terminal 30, attendees do not need to operate the attendee terminal 30 during the funeral.

[0051] (Second Embodiment) As shown in Figure 7, the distribution system 1 according to the second embodiment includes a scripture DB 18 (scripture database) in which the storage unit 13 stores scripture data. The scripture data includes modern Japanese translations of various Buddhist scriptures. The scriptures include, for example, the Amitabha Sutra, the Heart Sutra, and the Lotus Sutra. Unlike the first embodiment, when the control unit 12 of the distribution server 10 obtains the chanting text from the speech recognition processing unit 15, it recognizes the scripture that is currently being chanted and obtains the modern Japanese translation of that scripture from the scripture DB 18. The control unit 12 transmits the obtained modern Japanese translation of the scripture along with the chanting text to the natural language processing engine 16.

[0052] In the distribution system 1 according to the second embodiment, the distribution system 1 may function as a RAG (Retrieval-Augmented Generation) system. The control unit 12 instructs the natural language processing engine 16 in advance to generate a modern Japanese translation of the chanting text by referring to a modern Japanese translation of the scripture. When the control unit 12 recognizes a scripture in which chanting is being performed, it may send a modern Japanese translation of the entire scripture to the natural language processing engine 16. Alternatively, each time the control unit 12 sends a chanting text to the natural language processing engine 16, it may send the modern Japanese translation of the part of the scripture related to the chanting text to the natural language processing engine 16.

[0053] In the distribution system 1 according to the second embodiment, the process of step S206, which generates a modern Japanese translation of the chanting shown in Figure 4, differs from that of the first embodiment. An example of the process of step S206 in the second embodiment will be explained with reference to Figure 8.

[0054] First, the speech recognition processing unit 15 sends the chanting text to the distribution server 10 (step S401).

[0055] Next, the control unit 12 determines the scripture being chanted by the monk based on the chanting text obtained from the speech recognition processing unit 15, and obtains its modern Japanese translation from the scripture database 18 (step S402).

[0056] The control unit 12 inputs the chanting text and the modern Japanese translation of the related scriptures into the natural language processing engine 16 (step S403).

[0057] The natural language processing engine 16 translates the chanting text into modern Japanese by referring to the modern Japanese translation of the scriptures received from the distribution server 10 (step S404). If the monk chants faithfully according to the Buddhist scriptures, the natural language processing engine 16 sends the same content to the distribution server 10 as the modern Japanese translation of the scriptures sent from the distribution server 10. If the monk chants content different from that of the scriptures, the natural language processing engine 16 sends the distribution server 10 a modern Japanese translation that includes content different from the modern Japanese translation of the scriptures received from the distribution server 10, based on the content of the chanting. For example, if the monk repeats a part of the scriptures, the modern Japanese translation produced by the natural language processing engine 16 will reflect that repetition.

[0058] The natural language processing engine 16 sequentially sends the modern Japanese translation of the chanting text to the distribution server 10 (step S405).

[0059] As described above, the distribution system 1 can transmit a modern Japanese translation of the chanting by the monk to the participant terminal 30 in real time, similar to the first embodiment. Furthermore, the natural language processing engine 16 refers to a pre-prepared accurate modern Japanese translation of the scriptures to translate the chanting text into a modern Japanese translation, so it can always output a high-quality modern Japanese translation. Moreover, even if the monk's chanting is not accurately recognized by the speech recognition processing unit 15, the natural language processing engine 16 can refer to the modern Japanese translation of the scriptures received from the distribution server 10 to correct the speech recognition error and generate a correct modern Japanese translation.

[0060] While embodiments relating to this disclosure have been described based on the drawings and examples, it should be noted that those skilled in the art will find it easy to make various modifications or alterations based on this disclosure. Therefore, it should be noted that these modifications or alterations are included within the scope of this disclosure. For example, the functions included in each component or step can be rearranged in a logically consistent manner, and multiple components or steps can be combined into one or divided. Embodiments relating to this disclosure can also be realized as methods, programs, or storage media on which programs are recorded, executed by a processor in a device. These should also be understood to be included within the scope of this disclosure.

[0061] For example, although the above embodiment was described using a Buddhist funeral as an example, the distribution system and distribution method of this disclosure can also be applied to other religions in which the officiant recites scriptures written in a language other than modern language during the funeral. Also, for example, in the above embodiment, a QR code is read using a participant's terminal. However, the code information read by the participant's terminal is not limited to a QR code, but may be other two-dimensional codes or barcodes, etc. Furthermore, in the above embodiment, the speech recognition processing unit and the natural language processing engine were described as separate components. However, the speech recognition processing unit and the natural language processing engine may be implemented by a single AI system such as a multimodal AI system.

[0062] Some embodiments of the present disclosure are described below. However, it should be noted that the embodiments of the present disclosure are not limited to these. [Note 1] A distribution system that delivers a modern Japanese translation of Buddhist scriptures to the information terminals of attendees at a funeral, The aforementioned funeral includes an audio acquisition unit that sequentially acquires the audio of the chanting of sutras, A speech recognition processing unit that recognizes the audio of the chanting and outputs text data, A natural language processing engine that translates the aforementioned text data into the aforementioned modern Japanese translation, A distribution server having a control unit that sequentially distributes the modern Japanese translation obtained from the chanting of sutras to the information terminals of the attendees. A distribution system equipped with the following features. [Note 2] The distribution system as described in Appendix 1, wherein the control unit scrolls the modern Japanese translation displayed on the display screen of the information terminal and displays the translated modern Japanese translations sequentially at predetermined positions on the display screen. [Note 3] The distribution system as described in Appendix 2, wherein the control unit prohibits the participant from scrolling the display screen of the information terminal. [Note 4] The distribution system according to any one of the appendices 1 to 3, wherein when the control unit receives a processing request from the information terminal by reading code information installed in the funeral hall, it associates the information terminal with the distribution of the modern Japanese translation of a specific funeral based on the funeral schedule information included in the processing request. [Note 5] The distribution system according to any one of the appendices 1 to 4, wherein the speech recognition processing unit includes a speech recognition model generated by further training a pre-trained general-purpose speech recognition model using training data that includes pairs of chanting audio and text corresponding to the chanting audio. [Note 6] The distribution system described in any of the appendices 1 to 5, wherein the natural language processing engine, when it determines that the text data contains an error, outputs the modern Japanese translation with the error corrected. [Note 7] The distribution system according to any one of the appendices 1 to 6, wherein the distribution server comprises a storage unit for storing scripture data related to the chanting of sutras, the control unit inputs text data obtained from the speech recognition processing unit together with scripture data related to the text data to the natural language processing engine, and the natural language processing engine translates the text data into the modern Japanese translation based on the scripture data. [Note 8] The distribution system according to any one of the appendices 1 to 7, further comprising a speech synthesis processing unit that converts the aforementioned modern Japanese translation into spoken language, wherein the distribution server is configured to sequentially distribute the spoken language synthesized from the aforementioned modern Japanese translation, which has been translated from the chanting of sutras, to the information terminals of the attendees. [Note 9] A distribution method implemented by a distribution system that delivers a modern Japanese translation of Buddhist scriptures to the information terminals of attendees at a funeral, The audio of the chanting of sutras during the aforementioned funeral will be acquired sequentially, The system recognizes the audio of the chanting and outputs text data. Translating the aforementioned text data into the aforementioned modern Japanese translation, The modern Japanese translation derived from the aforementioned chanting will be sequentially distributed to the information terminals of the attendees. Distribution methods including those mentioned. [Explanation of symbols]

[0063] 1. Distribution System 10 Distribution Servers 11 Communications Department 12 Control Unit 13 Storage section 14 Input section 15. Speech Recognition Processing Unit 16 Natural Language Processing Engines 17. Speech Synthesis Processing Unit 18 Scripture Database 21. Microphone (sound acquisition unit) 22 Control section 23 Communication equipment 30. Attendee terminals (information terminals) 31 QR code (code information) 40 Networks 50 Funeral venues 51 Altar 52 monks' seats 53 Incense burner 54. Related to bereaved families 55 seats

Claims

1. A distribution system that delivers a modern Japanese translation of Buddhist scriptures to the information terminals of attendees at a funeral, The aforementioned funeral includes an audio acquisition unit that sequentially acquires the audio of the chanting of sutras, A speech recognition processing unit that recognizes the audio of the chanting and outputs text data, A natural language processing engine that translates the aforementioned text data into the aforementioned modern Japanese translation, A distribution server having a control unit that sequentially distributes the modern Japanese translation obtained from the chanting of sutras to the information terminals of the attendees. Equipped with, A distribution system in which, when the control unit receives a processing request from the information terminal by reading code information installed in the funeral hall, associates the information terminal with the distribution of the modern Japanese translation of a specific funeral based on the funeral schedule information included in the processing request.

2. The distribution system according to claim 1, wherein the control unit scrolls the modern Japanese translation displayed on the display screen of the information terminal and displays the translated modern Japanese translations sequentially at predetermined positions on the display screen.

3. The distribution system according to claim 2, wherein the control unit suppresses scrolling of the display screen of the information terminal by operation by the attendee.

4. The distribution system according to claim 1, wherein the speech recognition processing unit includes a speech recognition model generated by further training a pre-trained general-purpose speech recognition model using training data that includes a pair of chanting audio and text corresponding to the chanting audio.

5. The distribution system according to claim 1, wherein the natural language processing engine, when it determines that the text data contains an error, outputs the modern Japanese translation with the error corrected.

6. The distribution system according to claim 1, wherein the distribution server includes a storage unit for storing scripture data related to the chanting of sutras, the control unit inputs text data obtained from the speech recognition processing unit together with scripture data related to the text data to the natural language processing engine, and the natural language processing engine translates the text data into the modern Japanese translation based on the scripture data.

7. The distribution system according to claim 1, further comprising a speech synthesis processing unit that converts the modern Japanese translation into spoken language, wherein the distribution server is configured to sequentially distribute the spoken language synthesized from the modern Japanese translation translated from the chanting to the information terminals of the attendees.

8. A distribution method implemented by a distribution system that delivers a modern Japanese translation of Buddhist scriptures to the information terminals of attendees at a funeral, When the information terminal receives a processing request by reading code information installed within the funeral hall, the information terminal is associated with the distribution of the modern Japanese translation of a specific funeral based on the funeral schedule information included in the processing request. The audio of the chanting of sutras during the aforementioned funeral will be acquired sequentially, The system recognizes the audio of the chanting and outputs text data. Translating the aforementioned text data into the aforementioned modern Japanese translation, The modern Japanese translation derived from the aforementioned chanting will be sequentially distributed to the information terminals of the attendees. Distribution methods including those mentioned.

Images