Cloud robot group collaborative operation method and system based on voice authentication

Abstract

The invention discloses a cloud robot group collaborative operation method and system based on voice authentication. A cloud robot group collaborative operation method and system based on voice authentication have the beneficial effects that cloud robots are added according to network access information to form a cloud robot group, and a voice recognition technology can be applied to the cloud robot group under the technical problem that the voice recognition performance is reduced due to inconsistent environments. Based on a networking structure of the cloud robot group, a decision tree structure of the cloud robot group is determined, the decision tree structure of the cloud robot group can be based on cloud robots with different network access information, and the cloud robots can call cloud computing and cloud storage and integrate infrastructures and sharing services to provide services for the cloud robot group. And training the voice authentication model by using the cloud storage data of the cloud robot group to obtain the voice authentication model. And recognizing the received voice by using a voice authentication model and the cloud storage data of the cloud robot group.

Description

Technical Field

[0001] The present application relates to the field of voice communication algorithms, and more specifically, to a cloud robot group collaborative operation method and system based on voice authentication. Background Art

[0002] Speech recognition technology, also known as automatic speech recognition, aims to convert the vocabulary content in human speech into computer-readable input, such as keystrokes, binary codes, or character sequences. Speech recognition systems prompt customers to use new passwords in new situations, so that users do not need to remember fixed passwords and the system will not be deceived by recordings. Text-dependent speech recognition methods can be divided into dynamic time warping or hidden Markov model methods. Text-independent speech recognition has been studied for a long time, and the degradation of speech recognition performance caused by inconsistent environments is a major obstacle in applications.

[0003] How to apply speech recognition technology to cloud robot groups under the technical problem of decreased speech recognition performance caused by inconsistent environments is a dilemma of current voice communication algorithms. To this end, it is necessary to propose a cloud robot group collaborative operation method and system based on voice authentication to address the dilemma that traditional robot groups find it difficult to effectively use voice communication algorithms. Summary of the invention

[0004] The content of this application is used to introduce concepts in a brief form, which will be described in detail in the detailed implementation section below. The content of this application is not intended to identify the key features or essential features of the technical solution claimed for protection, nor is it intended to limit the scope of the technical solution claimed for protection.

[0005] In order to solve the technical problems mentioned in the above background technology section, some embodiments of the present application provide a cloud robot group collaborative operation method based on voice authentication, including: Add cloud robots according to network access information to form a cloud robot group; Based on the network structure of the cloud robot group, determine the decision tree structure of the cloud robot group; Determine whether the number of decision trees in the cloud robot group is greater than 1; If the number of decision trees of the cloud robot group is equal to the value 1, the cloud storage data of the cloud robot group is selected by random sampling; If the number of decision trees of the cloud robot group is greater than the value 1, the decision tree of the cloud robot group is selected by random sampling; Based on the decision tree of the selected cloud robot group, cloud storage data of the cloud robot group is selected by random sampling; Establish a model for voice authentication; Using the cloud storage data of the cloud robot group, the voice authentication model is trained to obtain a voice authentication model; The received voice is recognized using the voice authentication model and the cloud storage data of the cloud robot group.

[0006] Further, the cloud computing platform of the cloud robot is determined based on the network access information; Determine the machine learning library of the cloud computing platform of the cloud robot that is connected to the network; Determine the cloud storage platform of the cloud robot based on the network access information; Determine the storage data structure of the cloud storage of the cloud robot connected to the network.

[0007] Furthermore, the machine learning library based on the cloud computing platform matches the storage data structure of the cloud storage; Using the storage data structure of cloud storage, a training data set based on cloud storage is formed; The time sequence of calling the storage data structure of the cloud storage of each cloud robot connected to the network; Determine the data topology of the cloud robot group.

[0008] Furthermore, based on the data topology of the cloud robot group, the root node of the decision tree is determined; Using the data features of voice authentication, determine the internal nodes of the decision tree; Based on the result direction of voice authentication, determine the leaf nodes of the decision tree; The root node of the decision tree, the internal nodes of the decision tree and the leaf nodes of the decision tree are used as the decision tree structure of the cloud robot group.

[0009] Further, based on the time sequence of the storage data structure of the cloud storage, the flow direction of the data flow of the decision tree of the cloud robot group is determined; Using the flow direction of the data flow of the decision tree of the cloud robot group, determine the decision path from the root node of the decision tree to the internal node of the decision tree; Using the flow direction of the data flow of the decision tree of the cloud robot group, determine the decision path from the internal node of the decision tree to the leaf node of the decision tree; A decision tree structure of a cloud robot group with decision orientation is formed.

[0010] Furthermore, the decision tree structure of the cloud robot group is trained using the training data set based on cloud storage; Determine the decision success rate of the decision tree structure of the cloud robot group; Determine whether the decision success rate of the decision tree structure of the cloud robot group matches the evaluation criteria; If the decision success rate of the decision tree structure of the cloud robot group matches the evaluation criteria, then the number of decision trees of the cloud robot group is determined to be equal to the value 1; If the decision success rate of the decision tree structure of the cloud robot group does not match the evaluation criteria, it is determined that the number of decision trees of the cloud robot group is greater than a value of 1.

[0011] Further, the time sequence of the storage data structure of the cloud storage is called; Treat the time in the time series as a randomly selected value; Use the square median algorithm to select the time in the time series; The cloud storage data corresponding to the time in the time series is taken as the result of random sampling; The random sampling results are incorporated into the data training set of the voice authentication model.

[0012] Furthermore, the data received by the cloud robot is collected and preprocessed; Obtain preprocessed sample data; Extract features from sample data; Obtaining time domain characteristics and frequency domain characteristics of sample data; Based on the decision tree structure of the cloud robot group, the time domain features of the sample data are linked to the voice authentication result in a first manner, and the frequency domain features of the sample data are linked to the voice authentication result in a second manner; Based on the first connection and the second connection, forming a result judgment of a model of voice authentication; A model for speech authentication to be trained is formed.

[0013] Furthermore, the data training set of the voice authentication model is called; Training the model for speech authentication to be trained; Obtain the recognition accuracy of the voice authentication model; Determine whether the recognition accuracy of the voice authentication model is consistent with the evaluation index; If the recognition accuracy of the speech authentication model matches the evaluation index, the speech authentication model is output; If the recognition accuracy of the voice authentication model does not match the evaluation index, the data collection and preprocessing received by the cloud robot are returned until the voice authentication model is output.

[0014] Some embodiments of the present application also provide a cloud robot group collaborative operation system based on voice authentication, including: A server, used to execute the cloud robot group collaborative operation method based on voice authentication; All network-connected devices are connected to the server for communication.

[0015] In summary: According to the network access information, cloud robots are added to form cloud robot groups. The speech recognition technology can be applied to cloud robot groups under the technical problem of decreased speech recognition performance caused by inconsistent environments. Based on the network structure of the cloud robot group, the decision tree structure of the cloud robot group is determined. The decision tree structure of the cloud robot group can be based on cloud robots with different network access information. The cloud robot can call cloud computing and cloud storage, and integrate infrastructure and shared services to provide services for the cloud robot group. Different cloud computing and cloud storage can form different decision tree structures of cloud robot groups, thereby improving the speech recognition performance of cloud robot groups in inconsistent environments. Using at least one decision tree of a cloud robot group, a speech authentication model can be established. Using the cloud storage data of the cloud robot group, the speech authentication model is trained to obtain a speech authentication model. Using the speech authentication model and the cloud storage data of the cloud robot group, the received speech is recognized. The speech authentication model converts the vocabulary content in human speech into computer-readable input, such as keys, binary codes or character sequences. The speech recognition system prompts customers to use new passwords in new occasions, so that users do not need to remember fixed passwords and the system will not be deceived by recordings. Text-dependent sound recognition methods can be classified into dynamic time warping or hidden Markov model methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings constituting a part of this application are used to provide a further understanding of this application, so that other features, purposes and advantages of this application become more obvious. The illustrative embodiment drawings and their descriptions of this application are used to explain this application and do not constitute an improper limitation on this application.

[0017] In addition, throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that the components and elements are not necessarily drawn to scale.

[0018] In the attached picture: Figure 1 The present invention is a method flow chart of a cloud robot group collaborative operation method based on voice authentication.

[0019] Figure 2 It is a system structure connection diagram of a cloud robot group collaborative operation system based on voice authentication.

[0020] 100-server; 200-network access device. DETAILED DESCRIPTION

[0021] Embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms, and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are only for exemplary purposes and are not intended to limit the scope of protection of the present application.

[0022] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other.

[0023] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments.

[0024] Reference Figure 1 Some embodiments of the present application provide a cloud robot group collaborative operation method based on voice authentication, including: S100, adding a cloud robot according to the network access information to form a cloud robot group.

[0025] Specifically, the cloud computing platform of the cloud robot is determined based on the network access information. The machine learning library of the cloud computing platform of the cloud robot that has accessed the network is determined. The cloud storage platform of the cloud robot is determined based on the network access information. The storage data structure of the cloud storage of the cloud robot that has accessed the network is determined.

[0026] Based on the machine learning library of the cloud computing platform, match the storage data structure of the cloud storage. Use the storage data structure of the cloud storage to form a training data set based on the cloud storage. Call the time sequence of the storage data structure of the cloud storage of each cloud robot connected to the network. Determine the data topology structure of the cloud robot group.

[0027] In fact, cloud robot groups have the characteristics of knowledge sharing. Cloud robots can learn from each other and share knowledge through the Internet, solving the limitations of individual robots' self-learning. For example, RoboEarth is a website dedicated to robots, where robots can share information and learn from each other's behavior and environment.

[0028] Cloud robot groups can collaborate on business, and robots with different capabilities distributed around the world can cooperate efficiently, share information resources, and complete larger and more complex tasks.

[0029] The human-machine interaction characteristics of cloud robot groups. In practical applications, machines can provide better solutions by using interactive functions.

[0030] S200, determining a decision tree structure of the cloud robot group based on the network structure of the cloud robot group.

[0031] Specifically, based on the data topology of the cloud robot group, the root node of the decision tree is determined. The internal nodes of the decision tree are determined by using the data features of voice authentication. The leaf nodes of the decision tree are determined based on the result of voice authentication. The root node of the decision tree, the internal nodes of the decision tree, and the leaf nodes of the decision tree are used as the decision tree structure of the cloud robot group.

[0032] In fact, a tree-based model that builds a decision tree by conditionally judging data features can be used for classification and regression problems. Its advantages are strong interpretability and ability to handle nonlinear relationships, but its disadvantages are that it is prone to overfitting, especially when the data dimension is high and the sample size is small.

[0033] Based on the time sequence of the storage data structure of the cloud storage, the flow direction of the data flow of the decision tree of the cloud robot group is determined. The decision path from the root node of the decision tree to the internal node of the decision tree is determined by using the flow direction of the data flow of the decision tree of the cloud robot group. The decision path from the internal node of the decision tree to the leaf node of the decision tree is determined by using the flow direction of the data flow of the decision tree of the cloud robot group. A decision tree structure of the cloud robot group with decision orientation is formed.

[0034] As you can understand, the decision tree structure mainly consists of the following parts: nodes, branches, training data, and evaluation criteria.

[0035] The root node is the starting point of the decision tree and represents the beginning of the entire data set. In classification problems, it represents the test condition for a feature or attribute.

[0036] Internal nodes are intermediate nodes in a decision tree that represent a decision point or a feature test condition. Each internal node represents a test on an attribute that splits the data into two or more subsets.

[0037] The leaf node is the terminal node of the decision tree, which represents the final decision result or category label. In classification tasks, the value of the leaf node is usually the label of a certain category.

[0038] Lines extending from the root node or internal nodes represent the flow of data or decision paths. Branches are usually determined by the value of features or attributes and are used to divide data sets into smaller subsets.

[0039] Training data is the basis for building a decision tree, which usually contains input features and corresponding labels or output values. The decision tree generates corresponding decision rules and structures by learning and analyzing the training data.

[0040] Evaluation criteria are used to determine when to stop building a decision tree. Common evaluation criteria include entropy, Gini coefficient, etc. These criteria are used to measure the purity of a data set. When the purity of a data set reaches a certain threshold, the tree growth is stopped. For example, when the data under a node all belong to the same category, a state of high purity is reached, and the node may no longer be split.

[0041] S300, determining whether the number of decision trees of the cloud robot group is greater than a value of 1.

[0042] Specifically, the decision tree structure of the cloud robot group is trained using a training data set based on cloud storage. The decision success rate of the decision tree structure of the cloud robot group is determined. It is judged whether the decision success rate of the decision tree structure of the cloud robot group matches the evaluation standard. If the decision success rate of the decision tree structure of the cloud robot group matches the evaluation standard, it is determined that the number of decision trees of the cloud robot group is equal to the value 1. If the decision success rate of the decision tree structure of the cloud robot group does not match the evaluation standard, it is determined that the number of decision trees of the cloud robot group is greater than the value 1.

[0043] S400: If the number of decision trees of the cloud robot group is equal to the value 1, cloud storage data of the cloud robot group is selected by random sampling.

[0044] S500: If the number of decision trees of the cloud robot group is greater than the value 1, select the decision tree of the cloud robot group by random sampling.

[0045] S600, based on the decision tree of the selected cloud robot group, select cloud storage data of the cloud robot group by random sampling.

[0046] Specifically, call the time series of the storage data structure of the cloud storage. Use the time in the time series as a randomly selected value. Use the square median algorithm to select the time in the time series. Use the cloud storage data corresponding to the time in the time series as the result of random sampling. Incorporate the result of random sampling into the data training set of the voice authentication model.

[0047] S700: Establish a voice authentication model.

[0048] Specifically, the data received by the cloud robot is collected and preprocessed. Preprocessed sample data is obtained. Feature extraction is performed on the sample data. Time domain features and frequency domain features of the sample data are obtained. Based on the decision tree structure of the cloud robot group, the time domain features of the sample data are connected to the voice authentication result in a first connection, and the frequency domain features of the sample data are connected to the voice authentication result in a second connection. Based on the first connection and the second connection, a result judgment of a voice authentication model is formed. A voice authentication model to be trained is formed.

[0049] Call the data training set of the voice authentication model. Train the voice authentication model to be trained. Obtain the recognition accuracy of the voice authentication model. Determine whether the recognition accuracy of the voice authentication model is consistent with the evaluation index. If the recognition accuracy of the voice authentication model is consistent with the evaluation index, output the voice authentication model. If the recognition accuracy of the voice authentication model is inconsistent with the evaluation index, return to the data collection and preprocessing received by the cloud robot until the voice authentication model is output.

[0050] S800, using the cloud storage data of the cloud robot group to train the voice authentication model to obtain the voice authentication model.

[0051] S900, using the voice authentication model and the cloud storage data of the cloud robot group, recognize the received voice.

[0052] This embodiment relates to a cloud robot group collaborative operation method based on voice authentication. Cloud robots are added according to network access information to form a cloud robot group. The voice recognition technology can be applied to the cloud robot group under the technical problem of reduced voice recognition performance caused by inconsistent environments. Based on the networking structure of the cloud robot group, the decision tree structure of the cloud robot group is determined. The decision tree structure of the cloud robot group can be based on cloud robots with different network access information. The cloud robot can call cloud computing and cloud storage, and integrate infrastructure and shared services to provide services for the cloud robot group. Different cloud computing and cloud storage can form different decision tree structures of cloud robot groups, thereby improving the voice recognition performance of cloud robot groups in inconsistent environments. A voice authentication model can be established using at least one decision tree of a cloud robot group. The voice authentication model is trained using the cloud storage data of the cloud robot group to obtain a voice authentication model. The received voice is recognized using the voice authentication model and the cloud storage data of the cloud robot group. The voice authentication model converts the vocabulary content in human voice into computer-readable input, such as keys, binary codes, or character sequences. The speech recognition system prompts the customer to use a new password in a new situation, so that the user does not need to remember a fixed password and the system will not be deceived by the recording. Text-related voice recognition methods can be divided into dynamic time warping or hidden Markov model methods.

[0053] Reference Figure 2 Some embodiments of the present application provide a cloud robot group collaborative operation system based on voice authentication, including: The server 100 is used to execute the cloud robot group collaborative operation method based on voice authentication.

[0054] The network access devices 200 are all connected to the server 100 for communication.

[0055] The present embodiment relates to a cloud robot group collaborative operation system based on voice authentication. The server 100 adds cloud robots according to the network access information to form a cloud robot group, and can realize the application of voice recognition technology in the cloud robot group under the technical problem of reduced voice recognition performance caused by inconsistent environments. Based on the networking structure of the cloud robot group, the decision tree structure of the cloud robot group is determined. The decision tree structure of the cloud robot group can be based on cloud robots with different network access information. The cloud robot can call cloud computing and cloud storage, and integrate infrastructure and shared services to provide services for the cloud robot group. Different cloud computing and cloud storage can form different decision tree structures of cloud robot groups, thereby improving the voice recognition performance of cloud robot groups in inconsistent environments. Using the decision tree of at least one cloud robot group, a voice authentication model can be established. Using the cloud storage data of the cloud robot group, the voice authentication model is trained to obtain the voice authentication model. The network access device 200 uses the voice authentication model and the cloud storage data of the cloud robot group to recognize the received voice.

[0056] The above description is only some preferred embodiments of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present application is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above invention concept. For example, the above features are replaced with (but not limited to) technical features with similar functions disclosed in the embodiments of the present application to form a technical solution.

Claims

1. A cloud robot group collaborative operation method based on voice authentication, comprising: Add cloud robots according to network access information to form a cloud robot group; Based on the network structure of the cloud robot group, determine the decision tree structure of the cloud robot group; Determine whether the number of decision trees in the cloud robot group is greater than 1; If the number of decision trees of the cloud robot group is equal to the value 1, the cloud storage data of the cloud robot group is selected by random sampling; If the number of decision trees of the cloud robot group is greater than the value 1, the decision tree of the cloud robot group is selected by random sampling; Based on the decision tree of the selected cloud robot group, cloud storage data of the cloud robot group is selected by random sampling; Establish a model for voice authentication; Using the cloud storage data of the cloud robot group, the voice authentication model is trained to obtain a voice authentication model; The received voice is recognized using the voice authentication model and the cloud storage data of the cloud robot group.

2. The cloud robot group collaborative operation method based on voice authentication according to claim 1 is characterized in that: The step of adding a cloud robot according to the network access information to form a cloud robot group includes: Determine the cloud computing platform of the cloud robot based on the network access information; Determine the machine learning library of the cloud computing platform of the cloud robot that is connected to the network; Determine the cloud storage platform of the cloud robot based on the network access information; Determine the storage data structure of the cloud storage of the cloud robot connected to the network.

3. The cloud robot group collaborative operation method based on voice authentication according to claim 2 is characterized in that: The step of adding a cloud robot according to the network access information to form a cloud robot group also includes: A machine learning library based on a cloud computing platform that matches the storage data structure of cloud storage; Using the storage data structure of cloud storage, a training data set based on cloud storage is formed; The time sequence of calling the storage data structure of the cloud storage of each cloud robot connected to the network; Determine the data topology of the cloud robot group.

4. The cloud robot group collaborative operation method based on voice authentication according to claim 3 is characterized by: The step of determining the decision tree structure of the cloud robot group based on the network structure of the cloud robot group includes: Determine the root node of the decision tree based on the data topology of the cloud robot group; Using the data features of voice authentication, determine the internal nodes of the decision tree; Based on the result direction of voice authentication, determine the leaf nodes of the decision tree; The root node of the decision tree, the internal nodes of the decision tree and the leaf nodes of the decision tree are used as the decision tree structure of the cloud robot group.

5. The cloud robot group collaborative operation method based on voice authentication according to claim 4 is characterized in that: The determining of the decision tree structure of the cloud robot group based on the network structure of the cloud robot group also includes: Determine the direction of the data flow of the decision tree of the cloud robot group based on the time sequence of the storage data structure of the cloud storage; Using the flow direction of the data flow of the decision tree of the cloud robot group, determine the decision path from the root node of the decision tree to the internal node of the decision tree; Using the flow direction of the data flow of the decision tree of the cloud robot group, determine the decision path from the internal node of the decision tree to the leaf node of the decision tree; A decision tree structure of a cloud robot group with decision orientation is formed.

6. The cloud robot group collaborative operation method based on voice authentication according to claim 5 is characterized by: The determining whether the number of decision trees of the cloud robot group is greater than a value of 1 includes: Using cloud storage-based training data sets, the decision tree structure of the cloud robot group is trained; Determine the decision success rate of the decision tree structure of the cloud robot group; Determine whether the decision success rate of the decision tree structure of the cloud robot group matches the evaluation criteria; If the decision success rate of the decision tree structure of the cloud robot group matches the evaluation criteria, then the number of decision trees of the cloud robot group is determined to be equal to the value 1; If the decision success rate of the decision tree structure of the cloud robot group does not match the evaluation criteria, it is determined that the number of decision trees of the cloud robot group is greater than a value of 1.

7. The cloud robot group collaborative operation method based on voice authentication according to claim 6 is characterized by: The decision tree based on the selected cloud robot group, using random sampling to select cloud storage data of the cloud robot group, includes: The time sequence of calling the storage data structure of cloud storage; Treat the time in the time series as a randomly selected value; Use the square median algorithm to select the time in the time series; The cloud storage data corresponding to the time in the time series is taken as the result of random sampling; The random sampling results are incorporated into the data training set of the voice authentication model.

8. The cloud robot group collaborative operation method based on voice authentication according to claim 7 is characterized in that: The model for establishing voice authentication includes: Collect and pre-process data received by the cloud robot; Obtain preprocessed sample data; Extract features from sample data; Obtaining time domain characteristics and frequency domain characteristics of sample data; Based on the decision tree structure of the cloud robot group, the time domain features of the sample data are linked to the voice authentication result in a first manner, and the frequency domain features of the sample data are linked to the voice authentication result in a second manner; Based on the first connection and the second connection, forming a result judgment of a model of voice authentication; A model for speech authentication to be trained is formed.

9. The cloud robot group collaborative operation method based on voice authentication according to claim 8 is characterized in that: The model for establishing voice authentication also includes: Call the data training set of the voice authentication model; Training the model for speech authentication to be trained; Obtain the recognition accuracy of the voice authentication model; Determine whether the recognition accuracy of the voice authentication model is consistent with the evaluation index; If the recognition accuracy of the speech authentication model matches the evaluation index, the speech authentication model is output; If the recognition accuracy of the voice authentication model does not match the evaluation index, the data collection and preprocessing received by the cloud robot are returned until the voice authentication model is output.

10. A cloud robot group collaborative operation system based on voice authentication, comprising: A server, configured to execute the cloud robot group collaborative operation method based on voice authentication as described in any one of claims 1 to 9; All network-connected devices are connected to the server for communication.

Images