A multi-voice co-convergence anti-interference Bluetooth headset translation system
By adopting a multi-voice co-sumer anti-interference translation system in multi-user Bluetooth headsets, and using half-duplex channels and independent channels to process voice input, the problem of voice output interference in traditional Bluetooth headsets is solved, and the translation efficiency and accuracy are improved. And enhances the freedom of user interaction.
Patent Information
- Application Number
- CN202311243072.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-09-25
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2043-09-25
AI Technical Summary
Traditional multi-user Bluetooth headsets have interference when processing multiple voice outputs and multiple voice inputs, resulting in reduced translation efficiency and accuracy.
The multi-voice convergence anti-interference Bluetooth headset translation system is adopted, including a voice input module, a voice mode conversion identification module and a voice output module. Through a half-duplex channel, the channel mode is determined based on the number of voice input electrical signals and the requested output object. Independent channel transmission, avoid interference, and set a pause in the summary channel to delay output.
It improves translation efficiency and accuracy, avoids voice output interference, and enhances user interaction freedom.
Smart Images

Figure CN117292702B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of wireless translation, and in particular to a multi-voice co-convergence anti-interference Bluetooth headset translation system. Background Art
[0002] With the development of artificial intelligence technology, automatic translation technology has become more and more mature. The ability of machine translation has gradually approached that of human translation. Moreover, the translation speed is faster and free, which greatly facilitates cross-language communication. Common translation machines include translation headphones, which are in the form of dual headphones. When two users need to communicate, one earphone is left for themselves and the other earphone is given to the other person.
[0003] When more than two users use Bluetooth headsets for translation, the voice input and output of traditional multi-user Bluetooth headsets are carried out in a "one-to-many (single voice input and multiple voice outputs)" mode. However, in multi-person activities, there is not only a "one-to-many" mode, but also a "many-to-many (multiple voice outputs and multiple voice inputs)" situation. When using traditional multi-user Bluetooth headsets for communication, the voice outputs will interfere with each other, which will increase the translation time and cause translation errors.
[0004] Therefore, we propose a multi-voice co-convergence anti-interference Bluetooth headset translation system. Summary of the invention
[0005] The purpose of the present invention is to solve the problem in the prior art that traditional multi-user Bluetooth headsets will interfere with each other when processing multiple voice outputs and multiple voice inputs, resulting in reduced translation efficiency and reduced translation accuracy, and to propose a multi-voice co-convergence anti-interference Bluetooth headset translation system.
[0006] In order to achieve the above object, the present invention adopts the following technical solutions:
[0007] A multi-voice co-convergence anti-interference Bluetooth headset translation system, comprising a multi-voice anti-interference translation system, the multi-voice anti-interference translation system is composed of a voice input module, a voice mode conversion identification module, a voice translation module and a voice output module, and the multi-voice anti-interference translation system adopts half-duplex channel communication;
[0008] The voice input module adopts a common voice recorder, the voice mode conversion identification module is composed of a voice input identification module, a voice output identification module and a mode conversion module, the voice translation module is composed of a voice input temporary register and a multilingual translator, and the voice output module is composed of a one-to-one voice output module, a one-to-many voice output module, a many-to-many voice output module and a many-to-one voice output module.
[0009] Preferably, the internal signal processing steps of the multi-language anti-interference translation system are:
[0010] S1: Voice input, in which the user input voice is recorded into the input end of the multi-voice anti-interference translation system in the co-convergence voice recorder;
[0011] S2: Input temporary storage, temporarily storing the input voice through the voice input temporary storage, and transmitting the temporarily stored voice to the voice input identification module;
[0012] S3: Input identification, determining the voice output mode by the number of temporarily stored voice input electrical signals. When there is one voice input electrical signal and one output object, the voice output mode is a one-to-one mode; when there is one voice input electrical signal and multiple output objects, the voice output mode is a one-to-many mode; when there are multiple voice input electrical signals and multiple output objects, the voice output mode is a many-to-many mode; when there are multiple voice input electrical signals and one output object, the voice output mode is a many-to-one mode;
[0013] S4: Voice output. After the voice mode conversion identification module identifies and determines the voice input and output modes, the corresponding voice output module is selected to transmit the temporarily stored voice input to the corresponding voice output terminal.
[0014] Preferably, in steps S1 and S2, the multi-voice anti-interference translation system is internally provided with voice input terminals matching the number of users, the multi-voice anti-interference translation system is internally provided with voice input registers matching the number of users, and the multiple voice input registers are independent of each other.
[0015] Preferably, in step S3, the number of input objects is determined according to the number of electrical signals transmitted by the voice input temporary register, and the number of output objects is determined according to the number of requested output objects;
[0016] When the voice input register transmits one electrical signal, it is determined as a single voice input; when the voice input register transmits more than one electrical signal, it is determined as multiple voice inputs; when the requested output object is one or there are multiple requested objects and they are the same, it is determined as a single voice output; when there are multiple requested output objects and they are not the same, it is determined as multiple voice outputs.
[0017] Preferably, the half-duplex channels used by the one-to-one mode, one-to-many mode, many-to-many mode and many-to-one mode are all different. The one-to-one mode uses a dual-user channel, the one-to-many mode uses a speech channel, the many-to-many mode uses a multi-user channel, and the many-to-one mode uses an aggregate channel. The dual-user channel, speech channel and aggregate channel are independent channels, the multi-user channel is a combined channel, and the multi-user channel is a combination of the dual-user channel, the speech channel and the aggregate channel.
[0018] Preferably, the combination selection of the summary channel and speech channel of the multi-user channel is determined according to the requested output object. The requested output object of the summary channel is one, and the requested output objects of the speech channel are multiple. The dual-user channel combination selection is determined only when the requested output object is one and the voice input electrical signal of the requested output object is single.
[0019] Preferably, in step S3, the identification step is:
[0020] S31: Input identification, the number of electrical signals transmitted by the voice input temporary register is determined by the voice input identification module. When the number of electrical signals is one, it is determined as a single voice input. When the number of electrical signals is greater than one, it is determined as a multiple voice input, and the determination result is transmitted to the voice output identification module;
[0021] S32: output identification, determining the speech content transmitted by the speech input temporary storage, that is, the requested output object, and transmitting the determination result to the mode conversion module;
[0022] S33: Pattern output, outputting the pattern result identified by the voice input identification module and the voice output identification module to the corresponding voice output module.
[0023] Preferably, in step S33, when a dual-user channel and a speech channel are used, the voice output of the request object output end is performed synchronously, and when a summary channel is used, the voice output is delayed according to the deadline for the voice input buffer to be transmitted to the voice input identification module, and in the summary channel, a pauser is provided in the voice output module for pausing the voice output.
[0024] Preferably, when the voice output mode is determined to be a one-to-one mode, the mode conversion module uses a one-to-one voice output module to output voice; when it is a one-to-many mode, the mode conversion module uses a one-to-many voice output module to output voice; when it is a many-to-many mode, the mode conversion module uses a many-to-many voice output module to output voice; when it is a many-to-one mode, the mode conversion module uses a many-to-one voice output module to output voice.
[0025] Compared with the prior art, the present invention has the following beneficial effects:
[0026] 1. Aiming at the problem that traditional multi-user Bluetooth headsets in the prior art will interfere with each other when processing multiple voice outputs and multiple voice inputs, resulting in reduced translation efficiency and translation accuracy, the communication channel to be used is determined according to the number of voice input electrical signals and the requested output object, and the channels are independent, and the voice input content is transmitted to the voice output end accordingly, avoiding listening interference caused by multiple voice outputs in the same channel, thereby improving translation efficiency and translation accuracy.
[0027] 2. When using the summary channel to listen to voice input, the voice input is output with a delay according to the voice input deadline to avoid multiple voices being output at the same time and causing voice output interference. An output pauser is provided to allow input in the middle of voice output, thereby improving the accuracy of the voice output end while achieving the effect of enhancing the user's freedom of translation interaction. BRIEF DESCRIPTION OF THE DRAWINGS
[0028] Figure 1 This is a block diagram of the modules of a multi-voice anti-interference translation system of a multi-voice co-convergence anti-interference Bluetooth headset translation system proposed by the present invention;
[0029] Figure 2 This is a flow chart of internal signal processing of a multi-voice co-convergence anti-interference Bluetooth headset translation system proposed by the present invention;
[0030] Figure 3 This is a flow chart of the identification process of the voice mode conversion identification module of the multi-voice co-convergence anti-interference Bluetooth headset translation system proposed by the present invention. DETAILED DESCRIPTION
[0031] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all 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.
[0032] Reference Figure 1-3 A multi-voice co-convergence anti-interference Bluetooth headset translation system includes a voice anti-interference translation system, the multi-voice anti-interference translation system is composed of a voice input module, a voice mode conversion identification module, a voice translation module and a voice output module, and the multi-voice anti-interference translation system adopts half-duplex channel communication;
[0033] The voice input module adopts a common voice recorder, the voice mode conversion identification module consists of a voice input identification module, a voice output identification module and a mode conversion module, the voice translation module consists of a voice input register and a multilingual translator, and the voice output module consists of a one-to-one voice output module, a one-to-many voice output module, a many-to-many voice output module and a many-to-one voice output module.
[0034] Through the above technical scheme, the communication channel to be used is determined according to the number of voice input electrical signals and the requested output object, and the channels are independent, and the voice input content is transmitted to the voice output end accordingly, avoiding listening interference caused by multiple voice outputs in the same channel. Through a separately established channel, the voice output module is connected to the multi-language translator, thereby ensuring that the translation system is not interfered with and the accuracy of the translation is ensured.
[0035] Based on the above, the internal signal processing steps of the multi-voice anti-interference translation system are:
[0036] S1: Voice input, in which the user input voice is recorded into the input end of the multi-voice anti-interference translation system in the co-convergence voice recorder;
[0037] S2: Input temporary storage, temporarily storing the input voice through the voice input temporary storage, and transmitting the temporarily stored voice to the voice input identification module;
[0038] S3: Input identification, determining the voice output mode by the number of temporarily stored voice input electrical signals. When there is one voice input electrical signal and one output object, the voice output mode is a one-to-one mode; when there is one voice input electrical signal and multiple output objects, the voice output mode is a one-to-many mode; when there are multiple voice input electrical signals and multiple output objects, the voice output mode is a many-to-many mode; when there are multiple voice input electrical signals and one output object, the voice output mode is a many-to-one mode;
[0039] S4: Voice output. After the voice mode conversion identification module identifies and determines the voice input and output modes, the corresponding voice output module is selected to transmit the temporarily stored voice input to the corresponding voice output terminal.
[0040] Through the above technical solution, in steps S1 and S2, the multi-voice anti-interference translation system is internally provided with voice input terminals matching the number of users, the multi-voice anti-interference translation system is internally provided with voice input temporary registers matching the number of users, and the multiple voice input temporary registers are independent of each other;
[0041] Based on the above, further, in step S3, the number of input objects is determined according to the number of electrical signals transmitted by the voice input temporary register, and the number of output objects is determined according to the number of requested output objects;
[0042] When the voice input temporary register transmits one electrical signal, it is determined as a single voice input; when the voice input temporary register transmits more than one electrical signal, it is determined as multiple voice inputs; when the requested output object is one or there are multiple requested objects and they are the same, it is determined as a single voice output; when there are multiple requested output objects and they are not the same, it is determined as multiple voice outputs;
[0043] Based on the above, the determination process of the requested output object is as follows: number each voice output terminal, such as A, B, C, D, ..., and then input the number of the requested dialogue when requesting the output object. In the above, if there are multiple and the same requested output objects, the same number of voice output terminals are used for dialogue when multiple voice inputs are used.
[0044] Based on the above, the half-duplex channels used in the one-to-one mode, one-to-many mode, many-to-many mode and many-to-one mode are all different. The one-to-one mode uses a dual-user channel, the one-to-many mode uses a speech channel, the many-to-many mode uses a multi-user channel, and the many-to-one mode uses an aggregate channel. The dual-user channel, the speech channel and the aggregate channel are independent channels, the multi-user channel is a combined channel, and the multi-user channel is a combination of the dual-user channel, the speech channel and the aggregate channel.
[0045] Further based on the above, the summary channel and speech channel combination selection of the multi-user channel is determined according to the requested output object. The requested output object of the summary channel is one, and the requested output objects of the speech channel are multiple. The dual-user channel combination selection is determined only when the requested output object is one and the voice input electrical signal of this requested output object is single.
[0046] Specifically, the identification steps in step S3 are:
[0047] S31: Input identification, the number of electrical signals transmitted by the voice input temporary register is determined by the voice input identification module. When the number of electrical signals is one, it is determined as a single voice input. When the number of electrical signals is greater than one, it is determined as a multiple voice input, and the determination result is transmitted to the voice output identification module;
[0048] S32: output identification, determining the speech content transmitted by the speech input temporary storage, that is, the requested output object, and transmitting the determination result to the mode conversion module;
[0049] S33: Pattern output, outputting the pattern result identified by the voice input identification module and the voice output identification module to the corresponding voice output module.
[0050] Through the above technical solution, in step S33, when the dual-user channel and the speech channel are used, the voice output of the request object output end is performed synchronously, and when the summary channel is used, the voice output is transmitted to the voice input identification module according to the deadline of the voice input buffer, and the voice output of the request object output end is delayed, and in the summary channel, a pauser is provided in the voice output module for pausing the voice output.
[0051] Based on the above, when the voice output mode is determined to be a one-to-one mode, the mode conversion module uses a one-to-one voice output module to output voice; when it is a one-to-many mode, the mode conversion module uses a one-to-many voice output module to output voice; when it is a many-to-many mode, the mode conversion module uses a many-to-many voice output module to output voice; when it is a many-to-one mode, the mode conversion module uses a many-to-one voice output module to output voice.
[0052] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.
Claims
1. A multi-voice co-convergence anti-interference Bluetooth headset translation system, characterized in that: It includes a multi-voice anti-interference translation system, which is composed of a voice input module, a voice mode conversion identification module, a voice translation module and a voice output module, and the multi-voice anti-interference translation system adopts half-duplex channel communication; The voice input module adopts a common voice input device, the voice mode conversion identification module is composed of a voice input identification module, a voice output identification module and a mode conversion module, the voice translation module is composed of a voice input temporary storage and a multi-language translator, and the voice output module is composed of a one-to-one voice output module, a one-to-many voice output module, a many-to-many voice output module and a many-to-one voice output module; The internal signal processing steps of the multi-voice anti-interference translation system are as follows: S1: voice input, in which the user input voice is recorded into the input end of the multi-voice anti-interference translation system in a common voice recorder; S2: Input temporary storage, temporarily storing the input voice through the voice input temporary storage, and transmitting the temporarily stored voice to the voice input identification module; S3: Input identification, determining the voice output mode by the number of temporarily stored voice input electrical signals. When there is one voice input electrical signal and one output object, the voice output mode is a one-to-one mode; when there is one voice input electrical signal and multiple output objects, the voice output mode is a one-to-many mode; when there are multiple voice input electrical signals and multiple output objects, the voice output mode is a many-to-many mode; when there are multiple voice input electrical signals and one output object, the voice output mode is a many-to-one mode; S4: voice output, after the voice mode conversion identification module identifies and determines the voice input and output mode, selects the corresponding voice output module and transmits the temporarily stored voice input to the corresponding voice output terminal; When the voice output mode is determined to be a one-to-one mode, the mode conversion module uses the one-to-one voice output module to output voice; when it is a one-to-many mode, the mode conversion module uses the one-to-many voice output module to output voice; when it is a many-to-many mode, the mode conversion module uses the many-to-many voice output module to output voice; when it is a many-to-one mode, the mode conversion module uses the many-to-one voice output module to output voice.
2. A multi-voice co-convergence anti-interference Bluetooth headset translation system according to claim 1, characterized in that: In steps S1 and S2, the multi-voice anti-interference translation system is internally provided with voice input terminals matching the number of users, the multi-voice anti-interference translation system is internally provided with voice input temporary registers matching the number of users, and the multiple voice input temporary registers are independent of each other.
3. The multi-voice co-convergence anti-interference Bluetooth headset translation system according to claim 1, characterized in that: In step S3, the number of input objects is determined according to the number of electrical signals transmitted by the voice input temporary register, and the number of output objects is determined according to the number of requested output objects; When the voice input register transmits one electrical signal, it is determined as a single voice input; when the voice input register transmits more than one electrical signal, it is determined as multiple voice inputs; when the requested output object is one or there are multiple requested objects and they are the same, it is determined as a single voice output; when there are multiple requested output objects and they are not the same, it is determined as multiple voice outputs.
4. The multi-voice co-convergence anti-interference Bluetooth headset translation system according to claim 3 is characterized in that: The half-duplex channels used by the one-to-one mode, one-to-many mode, many-to-many mode and many-to-one mode are all different. The one-to-one mode uses a dual-user channel, the one-to-many mode uses a speech channel, the many-to-many mode uses a multi-user channel, and the many-to-one mode uses an aggregate channel. The dual-user channel, speech channel and aggregate channel are independent channels, the multi-user channel is a combined channel, and the multi-user channel is a combination of the dual-user channel, the speech channel and the aggregate channel.
5. The multi-voice co-convergence anti-interference Bluetooth headset translation system according to claim 4, characterized in that: The combination selection of the summary channel and the speech channel of the multi-user channel is determined according to the requested output object. The requested output object of the summary channel is one, and the requested output objects of the speech channel are multiple. The combination selection of the dual-user channel is determined only when the requested output object is one and the voice input electrical signal of the requested output object is single.
6. A multi-voice co-convergence anti-interference Bluetooth headset translation system according to claim 5, characterized in that: In step S3, the identification steps are: S31: Input identification, the number of electrical signals transmitted by the voice input temporary register is determined by the voice input identification module. When the number of electrical signals is one, it is determined as a single voice input. When the number of electrical signals is greater than one, it is determined as a multiple voice input, and the determination result is transmitted to the voice output identification module; S32: output identification, determining the speech content transmitted by the speech input temporary storage, that is, the requested output object, and transmitting the determination result to the mode conversion module; S33: Pattern output, outputting the pattern result identified by the voice input identification module and the voice output identification module to the corresponding voice output module.
7. The multi-voice co-convergence anti-interference Bluetooth headset translation system according to claim 6, characterized in that: In step S33, when the dual-user channel and the speech channel are used, the voice output of the request object output end is performed synchronously. When the summary channel is used, the voice output is delayed according to the deadline for the voice input buffer to be transmitted to the voice input identification module. In the summary channel, a pauser is provided in the voice output module for pausing the voice output.
