Mobile robot voice data control method based on large language model

Abstract

The invention relates to the field of mobile robot voice data control processing, in particular to a mobile robot voice data control method based on a large language model, and the method comprises the steps: collecting real-time voice data, and building a control feature database of a mobile robot; establishing a voice data control analysis model of the mobile robot based on a large language model according to the control feature database of the mobile robot; a voice data control result of the mobile robot is obtained by using the voice data control analysis model of the mobile robot, continuous optimization and adjustment can be performed along with input of more data and continuous learning of the model so as to adapt to new requirements and changes, and meanwhile, dual verification iteration of model training and model output is established in the overall scheme so as to improve the accuracy of the voice data control result. And accurate and stable output of a voice data control result is ensured.

Description

Technical Field

[0001] The invention relates to the field of mobile robot voice data control processing, and in particular to a mobile robot voice data control method based on a large language model. Background Art

[0002] The field of mobile robots has made significant progress in recent years, covering a variety of application areas from industrial automation to service robots. The control methods have also developed rapidly from basic pre-trial command settings to today's voice recognition control. Although voice data control technology has brought convenience, it also has some shortcomings that cannot be ignored. First, there may be a certain deviation between the data command input and the actual robot execution, which may cause inconvenience or even danger in an emergency. Secondly, the robot command analysis after data recognition also has various problems such as errors in interference. Summary of the invention

[0003] In view of the shortcomings of the prior art, the present invention provides a mobile robot voice data control method based on a large language model, which achieves efficient and accurate robot voice data control by establishing diversified analysis and processing models of different data features.

[0004] To achieve the above object, the present invention provides a mobile robot voice data control method based on a large language model, comprising: S1, collect real-time voice data to establish a control feature database for the mobile robot; S2. Establishing a voice data control analysis model for the mobile robot based on a large language model according to the control feature database of the mobile robot; S3. Obtaining a mobile robot voice data control result using the mobile robot voice data control analysis model.

[0005] Preferably, the collecting of real-time voice data to establish a control feature database of the mobile robot comprises: Collect real-time voice data; Acquire the signal frequency, signal time and signal capacity corresponding to the real-time voice data as basic data features of the real-time voice data; A control feature database of the mobile robot is obtained according to the basic data features of the real-time voice data.

[0006] Furthermore, acquiring a control feature database of a mobile robot according to the basic data features of the real-time voice data includes: Obtaining the command receiving frequency of the mobile robot according to the signal frequency corresponding to the basic data feature of the real-time voice data; Acquiring the command execution time of the mobile robot according to the signal time corresponding to the basic data feature of the real-time voice data; Obtaining a command execution action of the mobile robot according to the signal capacity corresponding to the basic data feature of the real-time voice data; The command receiving frequency, command execution time and command execution action of the mobile robot are used to divide the command execution characteristics of the mobile robot; The command execution characteristics and basic data characteristics of the mobile robot are used as a control characteristic database of the mobile robot.

[0007] Furthermore, establishing a voice data control analysis model of the mobile robot based on a large language model according to the control feature database of the mobile robot includes: S2-1, using the control feature database of the mobile robot to correspond to the basic data features and establish a basic voice data analysis model based on a large language model; S2-2, using the control feature database of the mobile robot corresponding to the instruction execution feature to establish a basic voice data control model based on the large language model; S2-3. Use the basic voice data analysis model and the basic voice data control model to perform overlapping verification processing to obtain the voice data control analysis model of the mobile robot.

[0008] Furthermore, establishing a basic voice data analysis model based on a large language model using the basic data features corresponding to the control feature database of the mobile robot includes: S2-1-1, using the signal frequency of the basic data features corresponding to the control feature database of the mobile robot to divide into a first data set; S2-1-2, using the signal time of the basic data features corresponding to the control feature database of the mobile robot to divide into a second data set; S2-1-3. Obtain corresponding historical signal frequencies according to the first data set to divide into a third data set; S2-1-4, obtaining a corresponding historical signal time according to the second data set to divide it into a fourth data set; S2-1-5, using the signal capacity of the basic data features corresponding to the control feature database of the mobile robot to divide into a first verification set; S2-1-6. Obtain the corresponding historical signal capacity according to the first verification set to divide into a second verification set; S2-1-7, using the first data set and the second data set as input, the corresponding relationship between the first data set and the second data set as output, and establishing a speech data analysis output model based on training of a large language model; S2-1-8, using the third data set and the fourth data set as input, and the corresponding relationship between the third data set and the fourth data set as output, training based on the large language model to establish a speech data analysis verification model; S2-1-9, determine whether the model output of the speech data analysis output model corresponds to the first verification set, if so, execute S2-1-10, otherwise, return to S2-1-7; S2-1-10. Determine whether the model output of the speech data analysis verification model corresponds to the second verification set. If so, use the speech data analysis output model and the speech data analysis verification model as the basic speech data analysis model. Otherwise, return to S2-1-8.

[0009] Furthermore, establishing a basic voice data control model based on a large language model using the control feature database of the mobile robot corresponding to the instruction execution feature includes: S2-2-1, using the command receiving frequency corresponding to the command execution feature of the control feature database of the mobile robot to divide into a fifth data set; S2-2-2, using the control feature database of the mobile robot to divide the instruction execution time corresponding to the instruction execution feature into a sixth data set; S2-2-3, dividing the instruction execution actions corresponding to the instruction execution features of the control feature database of the mobile robot into a seventh data set; S2-2-4. Using the fifth data set as input and the sixth and seventh data sets as output, a basic speech data control model is established through training based on a large language model.

[0010] Furthermore, the voice data control analysis model of the mobile robot is obtained by performing overlapping verification processing using the basic voice data analysis model and the basic voice data control model, including: S2-3-1, determine whether the model output of the speech data analysis output model of the basic speech data analysis model corresponds to the model output of the speech data analysis verification model, if so, execute S2-3-2, otherwise, return to S2-1-1; S2-3-2, determine whether the model output of the speech data analysis output model of the basic speech data analysis model corresponds to the model output of the basic speech data control model, if so, execute S2-3-3, otherwise, obtain the signal frequency and signal time corresponding to the fifth data set to update the first data set and the second data set respectively, and return to S2-1-7; S2-3-3. Determine whether the model output of the voice data analysis verification model of the basic voice data analysis model corresponds to the model output of the basic voice data control model. If so, use the current basic voice data analysis model and the basic voice data control model as the voice data control analysis model of the mobile robot. Otherwise, obtain the corresponding instruction receiving frequency according to the third data set and the fourth data set to update the fifth data set, and return to S2-2-4.

[0011] Furthermore, the mobile robot voice data control analysis model is used to obtain the mobile robot voice data control result, including: S3-1, using the voice data control analysis model of the mobile robot to obtain the voice data control analysis output result of the mobile robot; S3-2, performing operation verification processing according to the voice data control analysis output result of the mobile robot to obtain the voice data control result of the mobile robot.

[0012] Furthermore, obtaining the output result of the voice data control analysis of the mobile robot by using the voice data control analysis model of the mobile robot includes: Using the voice data control analysis model of the mobile robot, obtaining a model output of a voice data analysis output model of a basic voice data analysis model as an initial control output result; Using the voice data control analysis model of the mobile robot, obtaining a model output of a voice data analysis verification model of a basic voice data analysis model as an initial control verification result; Using the voice data control analysis model of the mobile robot, obtaining a model output of a basic voice data control model of a basic voice data analysis model as an initial execution verification result; The initial control output result, the initial control verification result and the initial execution verification result are used as the voice data control analysis output result of the mobile robot.

[0013] Furthermore, the mobile robot voice data control result obtained by performing operation verification processing according to the output result of the voice data control analysis of the mobile robot includes: S3-2-1, determine whether the voice data control analysis output result of the mobile robot corresponds to the initial control output result and the initial execution verification result, if so, use the initial control output result as the voice data control result of the mobile robot, otherwise, execute S3-2-2; S3-2-2, determine whether the voice data control analysis output result of the mobile robot corresponds to the initial control output result and the initial control verification result, if so, execute S3-2-3, otherwise, the voice data control of the mobile robot fails; S3-2-3. Determine whether the voice data control analysis output result of the mobile robot corresponds to the initial control verification result and the initial execution verification result. If so, use the initial control verification result as the voice data control result of the mobile robot. Otherwise, the voice data control of the mobile robot fails.

[0014] Compared with the closest prior art, the present invention has the following beneficial effects: The speech data feature analysis and classification model established using the large language model can quickly process and analyze huge speech data sets. The model significantly improves the speed of data processing through parallel computing and optimization algorithms. At the same time, it has strong adaptability and can easily cope with speech data in different fields and scenarios. With the input of more data and the continuous learning of the model, it can be continuously optimized and adjusted to adapt to new needs and changes. At the same time, a dual verification iteration of model training and model output is established in the overall solution to ensure the accurate and stable output of speech data control results. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 The present invention provides a flow chart of a method for controlling speech data of a mobile robot based on a large language model. DETAILED DESCRIPTION

[0016] The specific implementation modes of the present invention will be further described in detail below in conjunction with the accompanying drawings.

[0017] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution 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 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 creative work are within the scope of protection of the present invention.

[0018] Embodiment 1:

[0019] The present invention provides a mobile robot voice data control method based on a large language model, such as Figure 1 As shown, including: S1, collect real-time voice data to establish a control feature database for the mobile robot; S2. Establishing a voice data control analysis model for the mobile robot based on a large language model according to the control feature database of the mobile robot; S3. Obtaining a mobile robot voice data control result using the mobile robot voice data control analysis model.

[0020] S1 specifically includes: S1-1, collect real-time voice data; S1-2, obtaining the signal frequency, signal time and signal capacity corresponding to the real-time voice data as basic data features of the real-time voice data; S1-3. Acquire a control feature database of the mobile robot according to the basic data features of the real-time voice data.

[0021] S1-3 specifically includes: S1-3-1, obtaining the command receiving frequency of the mobile robot according to the signal frequency corresponding to the basic data characteristics of the real-time voice data; S1-3-2, obtaining the command execution time of the mobile robot according to the signal time corresponding to the basic data feature of the real-time voice data; S1-3-3, obtaining the command execution action of the mobile robot according to the signal capacity corresponding to the basic data characteristics of the real-time voice data; S1-3-4, using the command receiving frequency, command execution time and command execution action of the mobile robot to divide the command execution characteristics of the mobile robot; S1-3-5. Utilize the command execution characteristics and basic data characteristics of the mobile robot as the control characteristic database of the mobile robot.

[0022] S2 specifically includes: S2-1, using the control feature database of the mobile robot to correspond to the basic data features and establish a basic voice data analysis model based on a large language model; S2-2, using the control feature database of the mobile robot corresponding to the instruction execution feature to establish a basic voice data control model based on the large language model; S2-3. Use the basic voice data analysis model and the basic voice data control model to perform overlapping verification processing to obtain the voice data control analysis model of the mobile robot.

[0023] S2-1 specifically includes: S2-1-1, using the signal frequency of the basic data features corresponding to the control feature database of the mobile robot to divide into a first data set; S2-1-2, using the signal time of the basic data features corresponding to the control feature database of the mobile robot to divide into a second data set; S2-1-3. Obtain corresponding historical signal frequencies according to the first data set to divide into a third data set; S2-1-4, obtaining a corresponding historical signal time according to the second data set to divide it into a fourth data set; S2-1-5, using the signal capacity of the basic data features corresponding to the control feature database of the mobile robot to divide into a first verification set; S2-1-6. Obtain the corresponding historical signal capacity according to the first verification set to divide into a second verification set; S2-1-7, using the first data set and the second data set as input, the corresponding relationship between the first data set and the second data set as output, and establishing a speech data analysis output model based on training of a large language model; S2-1-8, using the third data set and the fourth data set as input, and the corresponding relationship between the third data set and the fourth data set as output, training based on the large language model to establish a speech data analysis verification model; S2-1-9, determine whether the model output of the speech data analysis output model corresponds to the first verification set, if so, execute S2-1-10, otherwise, return to S2-1-7; S2-1-10. Determine whether the model output of the speech data analysis verification model corresponds to the second verification set. If so, use the speech data analysis output model and the speech data analysis verification model as the basic speech data analysis model. Otherwise, return to S2-1-8.

[0024] S2-2 specifically includes: S2-2-1, using the control feature database of the mobile robot to divide the instruction receiving frequency corresponding to the instruction execution feature into a fifth data set; S2-2-2, using the control feature database of the mobile robot to divide the instruction execution time corresponding to the instruction execution feature into a sixth data set; S2-2-3, dividing the instruction execution actions corresponding to the instruction execution features of the control feature database of the mobile robot into a seventh data set; S2-2-4. Using the fifth data set as input and the sixth and seventh data sets as output, a basic speech data control model is established through training based on a large language model.

[0025] S2-3 specifically includes: S2-3-1, determine whether the model output of the speech data analysis output model of the basic speech data analysis model corresponds to the model output of the speech data analysis verification model, if so, execute S2-3-2, otherwise, return to S2-1-1; S2-3-2, determine whether the model output of the speech data analysis output model of the basic speech data analysis model corresponds to the model output of the basic speech data control model, if so, execute S2-3-3, otherwise, obtain the signal frequency and signal time corresponding to the fifth data set to update the first data set and the second data set respectively, and return to S2-1-7; S2-3-3. Determine whether the model output of the voice data analysis verification model of the basic voice data analysis model corresponds to the model output of the basic voice data control model. If so, use the current basic voice data analysis model and the basic voice data control model as the voice data control analysis model of the mobile robot. Otherwise, obtain the corresponding instruction receiving frequency according to the third data set and the fourth data set to update the fifth data set, and return to S2-2-4.

[0026] S3 specifically includes: S3-1, using the voice data control analysis model of the mobile robot to obtain the voice data control analysis output result of the mobile robot; S3-2, performing operation verification processing according to the voice data control analysis output result of the mobile robot to obtain the voice data control result of the mobile robot.

[0027] S3-1 specifically includes: S3-1-1, using the voice data control analysis model of the mobile robot to obtain the model output of the voice data analysis output model of the basic voice data analysis model as the initial control output result; S3-1-2, using the voice data control analysis model of the mobile robot to obtain the model output of the voice data analysis verification model of the basic voice data analysis model as the initial control verification result; S3-1-3, using the voice data control analysis model of the mobile robot to obtain the model output of the basic voice data control model of the basic voice data analysis model as the initial execution verification result; S3-1-4. Utilize the initial control output result, initial control verification result and initial execution verification result as the voice data control analysis output result of the mobile robot.

[0028] S3-2 specifically includes: S3-2-1, determine whether the voice data control analysis output result of the mobile robot corresponds to the initial control output result and the initial execution verification result, if so, use the initial control output result as the voice data control result of the mobile robot, otherwise, execute S3-2-2; S3-2-2, determine whether the voice data control analysis output result of the mobile robot corresponds to the initial control output result and the initial control verification result, if so, execute S3-2-3, otherwise, the voice data control of the mobile robot fails; S3-2-3. Determine whether the voice data control analysis output result of the mobile robot corresponds to the initial control verification result and the initial execution verification result. If so, use the initial control verification result as the voice data control result of the mobile robot. Otherwise, the voice data control of the mobile robot fails.

[0029] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0030] The present invention is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0031] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0032] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0033] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A mobile robot voice data control method based on a large language model, characterized in that: include: S1, collect real-time voice data to establish a control feature database for the mobile robot; S1-1, collect real-time voice data; S1-2, obtaining the signal frequency, signal time and signal capacity corresponding to the real-time voice data as basic data features of the real-time voice data; S1-3, obtaining a control feature database of the mobile robot according to the basic data features of the real-time voice data S2. Establishing a voice data control analysis model for the mobile robot based on a large language model according to the control feature database of the mobile robot; S2-1, using the control feature database of the mobile robot to correspond to the basic data features and establish a basic voice data analysis model based on a large language model; S2-2, using the control feature database of the mobile robot corresponding to the instruction execution feature to establish a basic voice data control model based on the large language model; S2-3, using the basic voice data analysis model and the basic voice data control model to perform overlapping verification processing to obtain a voice data control analysis model of the mobile robot; S3, using the voice data control analysis model of the mobile robot to obtain the voice data control result of the mobile robot; S3-1, using the voice data control analysis model of the mobile robot to obtain the voice data control analysis output result of the mobile robot; S3-2, performing operation verification processing according to the voice data control analysis output result of the mobile robot to obtain the voice data control result of the mobile robot.

2. A mobile robot voice data control method based on a large language model as claimed in claim 1, characterized in that: Acquiring a control feature database of a mobile robot according to the basic data features of the real-time voice data includes: Obtaining the command receiving frequency of the mobile robot according to the signal frequency corresponding to the basic data feature of the real-time voice data; Acquiring the command execution time of the mobile robot according to the signal time corresponding to the basic data feature of the real-time voice data; Obtaining a command execution action of the mobile robot according to the signal capacity corresponding to the basic data feature of the real-time voice data; The command receiving frequency, command execution time and command execution action of the mobile robot are used to divide the command execution characteristics of the mobile robot; The command execution characteristics and basic data characteristics of the mobile robot are used as a control characteristic database of the mobile robot.

3. A mobile robot voice data control method based on a large language model as claimed in claim 2, characterized in that: Establishing a basic speech data analysis model based on a large language model using the basic data features corresponding to the control feature database of the mobile robot includes: S2-1-1, using the signal frequency of the basic data features corresponding to the control feature database of the mobile robot to divide into a first data set; S2-1-2, using the signal time of the basic data features corresponding to the control feature database of the mobile robot to divide into a second data set; S2-1-3. Obtain corresponding historical signal frequencies according to the first data set to divide into a third data set; S2-1-4, obtaining a corresponding historical signal time according to the second data set to divide it into a fourth data set; S2-1-5, using the signal capacity of the basic data features corresponding to the control feature database of the mobile robot to divide into a first verification set; S2-1-6. Obtain the corresponding historical signal capacity according to the first verification set to divide into a second verification set; S2-1-7, using the first data set and the second data set as input, the corresponding relationship between the first data set and the second data set as output, and establishing a speech data analysis output model based on training of a large language model; S2-1-8, using the third data set and the fourth data set as input, and the corresponding relationship between the third data set and the fourth data set as output, training based on the large language model to establish a speech data analysis verification model; S2-1-9, determine whether the model output of the speech data analysis output model corresponds to the first verification set, if so, execute S2-1-10, otherwise, return to S2-1-7; S2-1-10. Determine whether the model output of the speech data analysis verification model corresponds to the second verification set. If so, use the speech data analysis output model and the speech data analysis verification model as the basic speech data analysis model. Otherwise, return to S2-1-8.

4. A method for controlling speech data of a mobile robot based on a large language model as claimed in claim 3, characterized in that: Using the control feature database of the mobile robot corresponding to the instruction execution feature to establish a basic voice data control model based on a large language model includes: S2-2-1, using the control feature database of the mobile robot to divide the instruction receiving frequency corresponding to the instruction execution feature into a fifth data set; S2-2-2, using the control feature database of the mobile robot to divide the instruction execution time corresponding to the instruction execution feature into a sixth data set; S2-2-3, dividing the instruction execution actions corresponding to the instruction execution features of the control feature database of the mobile robot into a seventh data set; S2-2-4. Using the fifth data set as input and the sixth and seventh data sets as output, a basic speech data control model is established through training based on a large language model.

5. A method for controlling speech data of a mobile robot based on a large language model as claimed in claim 4, characterized in that: The voice data control analysis model of the mobile robot obtained by performing overlapping verification processing using the basic voice data analysis model and the basic voice data control model includes: S2-3-1, determine whether the model output of the speech data analysis output model of the basic speech data analysis model corresponds to the model output of the speech data analysis verification model, if so, execute S2-3-2, otherwise, return to S2-1-1; S2-3-2, determine whether the model output of the speech data analysis output model of the basic speech data analysis model corresponds to the model output of the basic speech data control model, if so, execute S2-3-3, otherwise, obtain the signal frequency and signal time corresponding to the fifth data set to update the first data set and the second data set respectively, and return to S2-1-7; S2-3-3. Determine whether the model output of the voice data analysis verification model of the basic voice data analysis model corresponds to the model output of the basic voice data control model. If so, use the current basic voice data analysis model and the basic voice data control model as the voice data control analysis model of the mobile robot. Otherwise, obtain the corresponding instruction receiving frequency according to the third data set and the fourth data set to update the fifth data set, and return to S2-2-4.

6. A method for controlling mobile robot voice data based on a large language model as claimed in claim 5, characterized in that: The method of obtaining the output result of the voice data control analysis of the mobile robot by using the voice data control analysis model of the mobile robot includes: Using the voice data control analysis model of the mobile robot, obtaining a model output of a voice data analysis output model of a basic voice data analysis model as an initial control output result; Using the voice data control analysis model of the mobile robot, obtaining a model output of a voice data analysis verification model of a basic voice data analysis model as an initial control verification result; Using the voice data control analysis model of the mobile robot, obtaining a model output of a basic voice data control model of a basic voice data analysis model as an initial execution verification result; The initial control output result, the initial control verification result and the initial execution verification result are used as the voice data control analysis output result of the mobile robot.

7. A method for controlling speech data of a mobile robot based on a large language model as claimed in claim 6, characterized in that: The mobile robot voice data control result obtained by performing operation verification processing according to the voice data control analysis output result of the mobile robot includes: S3-2-1, determine whether the voice data control analysis output result of the mobile robot corresponds to the initial control output result and the initial execution verification result, if so, use the initial control output result as the voice data control result of the mobile robot, otherwise, execute S3-2-2; S3-2-2, determine whether the voice data control analysis output result of the mobile robot corresponds to the initial control output result and the initial control verification result, if so, execute S3-2-3, otherwise, the voice data control of the mobile robot fails; S3-2-3. Determine whether the voice data control analysis output result of the mobile robot corresponds to the initial control verification result and the initial execution verification result. If so, use the initial control verification result as the voice data control result of the mobile robot. Otherwise, the voice data control of the mobile robot fails.

Images