System and method for realizing AI function of common audio equipment

Abstract

The invention discloses a system and method for achieving an AI function through common audio equipment, the system comprises the audio equipment, a Bluetooth receiver and a processing terminal, the audio equipment captures voice of a local party and transmits voice data of the local party to the processing terminal through the Bluetooth receiver, and a third-party application sends received voice data of the other party to the processing terminal; the processing terminal performs AI processing on the voice data of the local party and the voice data of the opposite party, generates a first translation voice and a second translation voice according to a real-time translation selection instruction, and respectively sends the first translation voice and the second translation voice to the opposite party and the Bluetooth receiver; and the Bluetooth receiver distributes the second translated voice to audio equipment for playing. According to the invention, common audio equipment without an AI chip can realize an AI intelligent function, new audio equipment does not need to be purchased, the cost is reduced, and the waste of existing resources is avoided.

Description

Technical Field

[0001] The present invention relates to the field of AI intelligent audio technology, and in particular to a system and method for realizing AI functions in a common audio device. Background Art

[0002] With the development of artificial intelligence technology, many audio devices are equipped with AI chips, which can provide users with intelligent services such as real-time voice translation, meeting recording and analysis, significantly improving work efficiency. Take AI smart conference headphones as an example. Through its built-in AI chip, it can not only realize recording transcription, real-time translation and storage, but also cooperate with third-party APPs to automatically generate meeting summaries, extract key information and to-do items, etc. These functions greatly optimize the meeting experience.

[0003] However, most current audio devices do not have AI chips and cannot enjoy the above-mentioned intelligent services. In order to obtain these advanced AI functions, users often need to invest extra costs to replace new audio devices, which not only increases the economic burden, but also leads to a waste of existing resources. Summary of the invention

[0004] In view of the above-mentioned defects, the technical problem to be solved by the present invention is to provide a system and method for realizing AI functions in ordinary audio devices, so as to solve the problem that ordinary audio devices cannot satisfy users' use of AI intelligent functions because they do not have AI chips.

[0005] To this end, in a first aspect, the present application provides a method for a common audio device to implement an AI function, comprising the following steps:

[0006] The processing terminal runs a third-party application, and in response to the access operation of the Bluetooth receiver, automatically runs the AI ​​wake-up tool stored on the Bluetooth receiver, loads the audio virtual driver, and maps the Bluetooth receiver as an audio device to the third-party application on the processing terminal;

[0007] The audio device captures the voice of the party and transmits the voice data of the party to the processing terminal through the Bluetooth receiver. The third-party application sends the received voice data of the other party to the processing terminal.

[0008] The processing terminal performs AI processing on the voice data of the party and the other party, and performs corresponding operations according to whether the real-time translation selection instruction is received:

[0009] When receiving the real-time translation selection instruction, a first translation voice and a second translation voice are generated and sent to the processing terminal; the processing terminal sends the first translation voice to the third-party application, and the third-party application transmits it to the other party; the second translation voice is sent to the Bluetooth receiver, and the Bluetooth receiver sends the second translation voice to the audio device for playback; the first translation voice is the language used by the other party formed by translating the own voice, and the second translation voice is the language used by the own party formed by translating the other party's voice;

[0010] When no real-time translation selection instruction is received, the voice data of the party is transmitted to the third-party application, and then transmitted to the other party by the third-party application; the voice data of the other party is transmitted to the Bluetooth receiver, and the Bluetooth receiver sends the voice data of the other party to the audio device for playback.

[0011] Based on the above technical solution, this application connects ordinary audio devices to the processing terminal through a Bluetooth receiver, and uses the AI ​​computing power of the processing terminal locally or in the cloud to perform AI intelligent processing on the audio signal, save meeting minutes, generate meeting summaries, extract to-do items, and perform real-time voice translation, so that ordinary audio devices without AI chips can realize AI intelligent functions. There is no need to purchase new audio equipment, which reduces costs and avoids the waste of existing resources.

[0012] In the above technical solution, preferably, the AI ​​computing power used for AI processing is deployed on the processing terminal or the cloud. In this implementation, deploying the AI ​​computing power on the processing terminal can greatly improve confidentiality and can be used in an environment without a network; while deploying it on the cloud improves application flexibility and reduces deployment costs.

[0013] In the above technical solution, on the basis of real-time voice transmission, a real-time voice translation function is also provided, so that both parties can communicate in their own language, which is convenient to use.

[0014] In the above technical solution, preferably, the AI ​​processing includes at least one of speech-to-text, real-time speech translation, and intelligent analysis of meeting records. Therefore, the present application solution can provide a variety of AI applications by accessing different large models, further improving work efficiency and experience.

[0015] In the above technical solution, preferably, the voice data of the own party, the voice data of the other party and the result of AI processing are stored locally in the processing terminal or in the cloud.

[0016] In the above technical solution, preferably, the voice data of the local party, the voice data of the other party and the result of AI processing are stored on the Bluetooth connector.

[0017] In the above technical solution, preferably, the third-party application is a remote conference application or an instant communication application.

[0018] In a second aspect, the present application also provides a system for realizing AI functions in a common audio device, including an audio device, a Bluetooth receiver and a processing terminal.

[0019] One side of the Bluetooth receiver is connected to the audio device, and the other side is connected to the processing terminal; the Bluetooth receiver stores an AI wake-up tool, and the processing terminal runs a third-party application, and after the Bluetooth receiver is connected, the AI ​​wake-up tool is automatically run, the audio virtual driver is loaded, and the Bluetooth receiver is mapped as an audio device to the third-party application on the processing terminal;

[0020] The audio device captures the voice of the party and transmits the voice data of the party to the processing terminal through the Bluetooth receiver. The third-party application sends the received voice data of the other party to the processing terminal.

[0021] The processing terminal performs AI processing on the voice data of the party and the other party, and performs corresponding operations according to whether the real-time translation selection instruction is received:

[0022] When receiving the real-time translation selection instruction, a first translation voice and a second translation voice are generated and sent to the processing terminal; the processing terminal sends the first translation voice to the third-party application, and the third-party application transmits it to the other party; the second translation voice is sent to the Bluetooth receiver, and the Bluetooth receiver sends the second translation voice to the audio device for playback; the first translation voice is the language used by the other party formed by translating the own voice, and the second translation voice is the language used by the own party formed by translating the other party's voice;

[0023] When no real-time translation selection instruction is received, the voice data of the party is transmitted to the third-party application, and then transmitted to the other party by the third-party application; the voice data of the other party is transmitted to the Bluetooth receiver, and the Bluetooth receiver sends the voice data of the other party to the audio device for playback.

[0024] In the above technical solution, preferably, the AI ​​computing power used for AI processing is deployed in the processing terminal or the cloud.

[0025] In the above technical solution, preferably, the AI ​​processing includes at least one of speech-to-text, real-time speech translation, and intelligent analysis of meeting records.

[0026] In the above technical solution, preferably, the voice data of the own party, the voice data of the other party and the result of AI processing are stored locally in the processing terminal or in the cloud.

[0027] It can be seen from the above technical solutions that the system and method for realizing AI functions in ordinary audio devices provided by the present invention solve the problem that ordinary audio devices in the prior art cannot satisfy users' use of AI intelligent functions because they do not have AI chips. Compared with the prior art, the present invention has the following beneficial effects:

[0028] Ordinary audio devices are connected to the processing terminal via a Bluetooth receiver, and the AI ​​computing power of the processing terminal locally or in the cloud is used to perform AI intelligent processing on the audio signal to save meeting records, generate meeting summaries, extract to-do items, and perform real-time voice translation. This enables ordinary audio devices without AI chips to implement AI intelligent functions, eliminating the need to purchase new audio equipment, reducing costs, and avoiding the waste of existing resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces and describes the drawings required for use in the embodiments of the present invention or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0030] Figure 1 A schematic diagram of a system for implementing AI functions on a common audio device provided by an embodiment of the present invention;

[0031] Figure 2 A schematic diagram of the system architecture of the Bluetooth receiver in the present invention;

[0032] Figure 3 A flow chart of a method for implementing AI functions in a common audio device provided in Embodiment 1 of the present invention;

[0033] Figure 4 A flow chart of a method for implementing AI functions in a common audio device provided in Embodiment 2 of the present invention. DETAILED DESCRIPTION

[0034] Existing AI smart audio devices, such as AI smart headphones, are equipped with dedicated AI processor chips, which can realize corresponding AI functions, such as real-time translation, intelligent analysis and recording of conferences, etc. In contrast, many traditional Bluetooth headphones cannot enjoy the convenience brought by AI intelligence due to the lack of such AI chips.

[0035] To this end, this application connects an ordinary audio device to a processing terminal through a Bluetooth receiver, uses the automatically running AI wake-up tool built into the Bluetooth receiver to map the Bluetooth receiver as an audio device for a third-party application, and uses the AI ​​computing power of the local or cloud processing terminal to perform AI intelligent processing on the audio signal, save meeting minutes, generate meeting summaries, extract to-do items, and perform real-time voice translation, so that ordinary Bluetooth audio devices without AI chips can realize AI intelligent functions, without the need to purchase new audio equipment, reducing costs and avoiding waste of existing resources.

[0036] This application scheme overcomes the defect that traditional audio devices cannot enjoy intelligent services due to the lack of AI dedicated chips, achieves dual optimization of technology upgrade and cost-effectiveness, and meets users' needs for intelligent experience.

[0037] The following will be combined with the accompanying drawings of the embodiments of the present invention to clearly and completely describe the technical solutions of the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0038] In order to more clearly explain and illustrate the technical solution and implementation of the present invention, several preferred specific embodiments for implementing the technical solution of the present invention are introduced below.

[0039] It should be noted that the directional words such as "inside, outside", "front, back" and "left, right" in this article are expressed based on the product usage status. Obviously, the use of the corresponding directional words does not constitute a limitation on the protection scope of this scheme.

[0040] See also Figure 1 , Figure 1 A system diagram of a common audio device implementing AI functions provided in this application.

[0041] like Figure 1 As shown, the present application provides a system for implementing AI functions in a common audio device, including an audio device 10, a Bluetooth receiver 20 and a processing terminal 30.

[0042] The audio device 10 is an ordinary Bluetooth audio device without an AI chip, and cannot provide AI intelligent services such as real-time recording, transcription, voice translation, meeting record and analysis. The audio device 10 has a microphone and a speaker. The microphone is used to collect the user's own voice, and the speaker is used to play the received voice of the other party.

[0043] like Figure 2As shown, the Bluetooth receiver 20 includes a processor 1 , a memory 2 , a storage 3 , a Bluetooth module 4 , a communication interface 5 , a button 6 and an indicator light 7 .

[0044] Among them, processor 1 and memory 2 constitute the minimum computing system to realize system functions.

[0045] The memory 3 is connected to the processor 1 and stores an automatically running AI wake-up tool thereon. The AI ​​wake-up tool can be used to realize functions such as virtual mapping of audio devices and data transmission between the audio device 10 and the processing terminal 30.

[0046] Specifically, the AI ​​wake-up tool includes loading an audio virtual driver module and a data transmission module. The automatic operation of the AI ​​wake-up tool is a prior art, for example, an autorun.inf configuration file is set in the root directory of the memory 3, and the saved AI wake-up tool is automatically operated according to the specification in the file. That is, when the Bluetooth receiver 20 is inserted into the processing terminal 30, the processing terminal 30 automatically operates the AI ​​wake-up tool.

[0047] The Bluetooth module 4 is used for pairing and connecting with an ordinary Bluetooth audio device, receiving the local voice picked up by the ordinary Bluetooth audio device through the Bluetooth module, and sending the received other party's voice to the ordinary Bluetooth audio device for playback.

[0048] The communication interface 5 is used to connect to the processing terminal 30 to achieve data interaction between the two. In one embodiment, the communication interface 5 can use Type-C, for example, to connect to the processing terminal via a data cable.

[0049] Button 6 is used to activate corresponding functions, such as power on / off, recording, real-time translation, etc.

[0050] The LED indicator light 7 is used to indicate the corresponding working status of the Bluetooth receiver.

[0051] The above-mentioned Bluetooth receiver is connected to the audio device 10 through the Bluetooth module and connected to the processing terminal 30 through the communication interface. The Bluetooth receiver 20 sends the local voice collected by the audio device 10 to the processing terminal 30, and sends the other party's voice or translated voice received by the processing terminal 30 to the audio device 10 for playback.

[0052] The processing terminal 30 is installed with a third-party application, which may be a conference application or an instant communication application, such as Tencent Conference, Feishu, Teams, DingTalk, Telegram, etc.

[0053] The processing terminal 30 runs a third-party application on the processing terminal 30, and in response to the access operation of the Bluetooth receiver 20, automatically runs the AI ​​wake-up tool stored on the Bluetooth receiver 20 and runs it in the background, loads the audio virtual driver, and maps the Bluetooth receiver 20 as an audio device to the third-party application on the processing terminal 30.

[0054] The above is the operating mode in which the processing terminal 30 first runs the third-party application and then inserts the Bluetooth receiver 20. If the Bluetooth receiver 20 is inserted first, the operating system of the processing terminal 30 will detect the automatically running AI wake-up tool, and will first run the AI ​​wake-up tool and load the audio virtual driver. The AI ​​wake-up tool will run in the background. When the processing terminal 30 runs the third-party application, the AI ​​wake-up tool will map the Bluetooth receiver 20 as an audio device to the third application.

[0055] After the third-party application is started, during communication, the audio device 10 captures the voice of the party and transmits the voice data of the party to the processing terminal 30 through the Bluetooth receiver 20 , and the third-party application sends the received voice data of the other party to the processing terminal 30 .

[0056] The processing terminal 30 performs AI processing on the voice data of the party and the voice data of the other party, including at least one of speech-to-text, real-time speech translation, and intelligent analysis of meeting records. The specific functions can be set according to the needs of the user and are not limited here. For example, different large models can be connected to realize the corresponding AI functions. The AI ​​computing power used for AI processing is deployed in the processing terminal or the cloud.

[0057] In this process, in response to the real-time translation selection instruction, the AI ​​computing power can generate a first translation voice and a second translation voice, and send them to the processing terminal. The first translation voice is the language used by the other party formed by the translation of the own voice, and the second translation voice is the language used by the own party formed by the translation of the other party's voice.

[0058] The processing terminal 30 sends the first translated voice to the third-party application, and the third-party application transmits it to the other party; at the same time, the processing terminal 30 sends the second translated voice to the Bluetooth receiver 20, and the Bluetooth receiver 20 sends the second translated voice to the audio device 10 for playback.

[0059] During this process, if the processing terminal 30 does not receive the real-time translation selection instruction, the voice data of the party will be transmitted to the third-party application, and the third-party application will transmit it to the other party; the voice data of the other party will be transmitted to the Bluetooth receiver 20, and the Bluetooth receiver 20 will send the voice data of the other party to the audio device 10 for playback.

[0060] As a result, real-time voice communication and real-time voice translation are realized. Moreover, after AI processing, it can automatically generate meeting summaries, extract key information and to-do items, etc., which improves efficiency and is easy to use.

[0061] In this application, the AI ​​computing power (AI computing power module) that implements AI functions can be provided by local or cloud deployment.

[0062] Compared with the local deployment of AI computing power, cloud deployment solutions have demonstrated significant flexibility and wide applicability. For example, they can provide services for multiple conference scenarios at the same time, adapt to changing business needs, and reduce hardware investment and maintenance costs.

[0063] Based on the above system, the present application also provides a method for implementing AI functions in a common audio device, including the following steps:

[0064] The processing terminal runs a third-party application, and in response to the Bluetooth receiver access operation, automatically runs the AI ​​wake-up tool stored on the Bluetooth receiver, loads the audio virtual driver, and maps the Bluetooth receiver as an audio device to the third-party application on the processing terminal;

[0065] The audio device captures the voice of the party and transmits the voice data of the party to the processing terminal through the Bluetooth receiver. The third-party application sends the received voice data of the other party to the processing terminal.

[0066] The processing terminal performs AI processing on the voice data of the party and the voice data of the other party, and generates a first translated voice and a second translated voice in response to the real-time translation selection instruction, and sends them to the processing terminal; the first translated voice is the language used by the other party formed by translating the voice of the party, and the second translated voice is the language used by the party formed by translating the voice of the other party;

[0067] The processing terminal sends the first translated voice to the third-party application, which is then transmitted to the other party by the third-party application; the processing terminal sends the second translated voice to the Bluetooth receiver, which sends the second translated voice to the audio device for playback; when the processing terminal does not receive the real-time translation selection instruction, the voice data of the party is transmitted to the third-party application, which is then transmitted to the other party by the third-party application; the voice data of the other party is transmitted to the Bluetooth receiver, which sends the voice data of the other party to the audio device for playback.

[0068] The following uses an ordinary conference omnidirectional microphone as an example to specifically explain the method for realizing AI functions in ordinary audio devices provided by this application.

[0069] See also Figure 3 , Figure 3A flow chart of a method for implementing AI functions in a common audio device provided in Example 1 of the present application, wherein the AI ​​functions in this embodiment include speech transcription, real-time speech translation and intelligent analysis.

[0070] like Figure 3 As shown, a method for implementing an AI function of a common audio device provided in Embodiment 1 is applied to a user initiating end, and the other end of the conversation with the user is called the other party. The method includes the following steps:

[0071] Step 110, the processing terminal runs a third-party application, and in response to the access operation of the Bluetooth receiver, automatically runs the AI ​​wake-up tool stored on the Bluetooth receiver, loads the audio virtual driver, and maps the Bluetooth receiver as an audio device to the third-party application on the processing terminal.

[0072] Step 120: the audio device captures the voice of the party concerned and transmits the voice data of the party concerned to the processing terminal via the Bluetooth receiver. The third-party application sends the received voice data of the other party to the processing terminal.

[0073] Step 130: The processing terminal uses local or cloud AI computing power to perform AI processing on the voice data of the party concerned and the voice data of the other party concerned, and generates a first translated voice and a second translated voice in response to a real-time translation selection instruction, and sends them to the processing terminal.

[0074] Specifically, the processing terminal uses local or cloud AI computing power to perform real-time AI processing on the party's voice data and the other party's voice data, uses the sound-to-text function to convert the party's voice data and the other party's voice data into text, and generates a first translated voice and a second translated voice respectively, wherein the first translated voice is the language used by the other party formed by the translation of the party's voice, and the second translated voice is the language used by the party formed by the translation of the other party's voice.

[0075] Step 140: the processing terminal sends the first translated voice to the third-party application, and the third-party application transmits it to the other party; the second translated voice is sent to the Bluetooth receiver, and the Bluetooth receiver sends the second translated voice to the audio device for playback.

[0076] After step 140 is completed, the process goes to step 120 and continues until the call ends.

[0077] The real-time translation function can be achieved through the button on the Bluetooth receiver or the function button on the third-party application.

[0078] See also Figure 4 , Figure 4 A flow chart of a method for implementing AI functions in a common audio device provided in Example 2 of the present application, in which the AI ​​functions include speech transcription and intelligent analysis, but do not include real-time speech translation.

[0079] like Figure 4 As shown, a method for implementing an AI function of a common audio device provided in Embodiment 2 is applied to a user initiating end, and the other end of the conversation with the user is called the other party. The method includes the following steps:

[0080] Step 210, the processing terminal runs a third-party application, and in response to the access operation of the Bluetooth receiver, automatically runs the AI ​​wake-up tool stored on the Bluetooth receiver, loads the audio virtual driver, and maps the Bluetooth receiver as an audio device to the third-party application on the processing terminal.

[0081] Step 220: the audio device captures the voice of the party concerned and transmits the voice data of the party concerned to the processing terminal via the Bluetooth receiver. The third-party application sends the received voice data of the other party to the processing terminal.

[0082] Step 230: The processing terminal uses local or cloud AI computing power to perform AI processing on the voice data of the party itself and the voice data of the other party.

[0083] Step 240, the processing terminal sends the voice of the party to the third-party application, and the third-party application transmits it to the other party; the processing terminal sends the voice of the other party to the Bluetooth receiver, and the Bluetooth receiver sends the voice of the other party to the audio device for playback.

[0084] In this application, the AI ​​processing results include meeting records and intelligent analysis results such as key information and to-do items, and can generate meeting summaries and minutes after the meeting.

[0085] In the above steps, the voice data and AI processing results can be saved locally on the processing terminal, in the cloud, or on a Bluetooth receiver.

[0086] In the present application method, through the transfer and mapping of the Bluetooth receiver, ordinary Bluetooth audio devices that lack AI dedicated chips can use AI intelligent services without purchasing new AI audio devices. The AI ​​function is realized at a relatively low cost, which broadens the application scope of existing ordinary Bluetooth audio devices.

[0087] Based on the description of the above specific embodiments, the system and method for realizing AI function of common audio equipment provided by the present invention have the following advantages compared with the prior art:

[0088] First, through the transfer and mapping of the Bluetooth receiver, ordinary Bluetooth audio devices that lack AI dedicated chips can also use AI intelligent services, which improves work efficiency and eliminates the need to purchase new AI audio devices, which is low-cost.

[0089] Second, AI computing power can be deployed locally or in the cloud to meet the needs of different users. For example, those with high confidentiality requirements can be deployed locally, while those with high performance requirements can be deployed in the cloud.

[0090] Finally, it should be noted that the terms "comprises", "includes" or any other variants thereof as used herein are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

[0091] The present invention is not limited to the above-mentioned optimal implementation mode. Anyone should be aware that any structural changes made under the inspiration of the present invention, and any technical solutions that are the same or similar to the present invention, fall within the protection scope of the present invention.

Claims

1. A method for realizing AI function in a common audio device, characterized in that: The following steps are involved: The processing terminal runs a third-party application, and in response to the access operation of the Bluetooth receiver, automatically runs the AI ​​wake-up tool stored on the Bluetooth receiver, loads the audio virtual driver, and maps the Bluetooth receiver as an audio device to the third-party application on the processing terminal; The audio device captures the voice of the party and transmits the voice data of the party to the processing terminal through the Bluetooth receiver. The third-party application sends the received voice data of the other party to the processing terminal. The processing terminal performs AI processing on the voice data of the party and the other party, and performs corresponding operations according to whether the real-time translation selection instruction is received: When receiving the real-time translation selection instruction, a first translation voice and a second translation voice are generated and sent to the processing terminal; the processing terminal sends the first translation voice to the third-party application, and the third-party application transmits it to the other party; the second translation voice is sent to the Bluetooth receiver, and the Bluetooth receiver sends the second translation voice to the audio device for playback; the first translation voice is the language used by the other party formed by translating the own voice, and the second translation voice is the language used by the own party formed by translating the other party's voice; When no real-time translation selection instruction is received, the voice data of the party is transmitted to the third-party application, and then transmitted to the other party by the third-party application; the voice data of the other party is transmitted to the Bluetooth receiver, and the Bluetooth receiver sends the voice data of the other party to the audio device for playback.

2. The method according to claim 1, characterized in that The AI ​​computing power used for AI processing is deployed locally on the processing terminal or in the cloud.

3. The method according to claim 1, characterized in that: The AI ​​processing includes at least one of speech-to-text, real-time speech translation, and intelligent analysis of meeting records.

4. The method according to claim 1, characterized in that: The voice data of the party concerned, the voice data of the other party concerned and the results of AI processing are stored locally in the processing terminal or in the cloud.

5. The method according to claim 1, characterized in that The local voice data, the other party's voice data and the result of AI processing are stored on the Bluetooth connector.

6. The method according to claim 1, characterized in that The third-party application is a remote conference application or an instant communication application.

7. A system for realizing AI function in common audio equipment, characterized in that: Includes audio equipment, Bluetooth receiver and processing terminal; One side of the Bluetooth receiver is connected to the audio device, and the other side is connected to the processing terminal; the Bluetooth receiver stores an AI wake-up tool, and the processing terminal runs a third-party application, and after the Bluetooth receiver is connected, the AI ​​wake-up tool is automatically run, the audio virtual driver is loaded, and the Bluetooth receiver is mapped as an audio device to the third-party application on the processing terminal; The audio device captures the voice of the party and transmits the voice data of the party to the processing terminal through the Bluetooth receiver. The third-party application sends the received voice data of the other party to the processing terminal. The processing terminal performs AI processing on the voice data of the party and the other party, and performs corresponding operations according to whether the real-time translation selection instruction is received: When receiving the real-time translation selection instruction, a first translation voice and a second translation voice are generated and sent to the processing terminal; the processing terminal sends the first translation voice to the third-party application, and the third-party application transmits it to the other party; the second translation voice is sent to the Bluetooth receiver, and the Bluetooth receiver sends the second translation voice to the audio device for playback; the first translation voice is the language used by the other party formed by translating the own voice, and the second translation voice is the language used by the own party formed by translating the other party's voice; When no real-time translation selection instruction is received, the voice data of the party is transmitted to the third-party application, and then transmitted to the other party by the third-party application; the voice data of the other party is transmitted to the Bluetooth receiver, and the Bluetooth receiver sends the voice data of the other party to the audio device for playback.

8. The system according to claim 7, characterized in that The AI ​​computing power used for AI processing is deployed on the processing terminal or in the cloud.

9. The system according to claim 7, characterized in that The AI ​​processing includes at least one of speech-to-text, real-time speech translation, and intelligent analysis of meeting records.

10. The system according to claim 7, characterized in that The voice data of the party concerned, the voice data of the other party concerned and the results of AI processing are stored locally in the processing terminal or in the cloud.

Images