阅读说明：本技术 语音数据的处理方法及装置、存储介质、电子装置 (Voice data processing method and device, storage medium and electronic device ) 是由 朱文博 于 2021-06-30 设计创作，主要内容包括：本发明提供了一种语音数据的处理方法及装置、存储介质、电子装置,上述方法包括：获取待处理的语音数据；根据多个预设语音模型中各预设语音模型对应的权重,从多个预设语音模型中确定至少一个目标语音模型,各预设语音模型的权重表征该预设语音模型识别结果的置信度；通过至少一个目标语音模型对待处理的语音数据进行处理,解决了现有技术中在使用多种语音识别引擎(即语音模型)进行语音识别时,识别时间长,无法确定识别结果的准确率等问题,确保了语音数据进行识别的灵活性,提升对于识别准确率的确定时间。(The invention provides a method and a device for processing voice data, a storage medium and an electronic device, wherein the method comprises the following steps: acquiring voice data to be processed; determining at least one target voice model from a plurality of preset voice models according to the weight corresponding to each preset voice model in the plurality of preset voice models, wherein the weight of each preset voice model represents the confidence coefficient of the recognition result of the preset voice model; the voice data to be processed is processed through the at least one target voice model, the problems that in the prior art, when multiple voice recognition engines (namely voice models) are used for voice recognition, recognition time is long, the accuracy of recognition results cannot be determined and the like are solved, the flexibility of voice data recognition is ensured, and the determination time of the recognition accuracy is improved.)

1. A method for processing voice data, comprising:

acquiring voice data to be processed;

determining at least one target voice model from a plurality of preset voice models according to the weight corresponding to each preset voice model in the preset voice models, wherein the weight of each preset voice model represents the confidence coefficient of the recognition result of the preset voice model;

and processing the voice data to be processed through the at least one target voice model.

2. The method of claim 1, wherein prior to obtaining the voice data to be processed, the method further comprises:

obtaining sample voice for training the preset voice models;

respectively processing the sample voice through the plurality of preset voice models to obtain a recognition result and a confidence degree corresponding to each preset voice model;

and determining weights corresponding to the preset voice models according to the recognition results and the confidence degrees corresponding to the preset voice models.

3. The method according to claim 2, wherein the processing the sample speech through the plurality of preset speech models respectively to obtain the recognition result corresponding to each preset speech model comprises:

obtaining standard identification data of the sample voice, wherein the standard identification data is used for indicating the sample voice to correctly analyze corresponding text content;

determining the difference between the standard recognition data and the recognition data obtained by processing the sample voice by each preset voice model;

and determining the recognition result of each preset voice model for the sample voice according to the difference.

4. The method of claim 2, wherein the step of processing the sample speech through the plurality of preset speech models to obtain confidence levels corresponding to the preset speech models comprises:

obtaining a confidence interval corresponding to the sample voice;

determining the probability of the recognition value of each preset voice model for the sample voice processing and the confidence interval, wherein the recognition value is used for indicating the number of repeated words of the recognition data of each preset voice model for the sample voice recognition and the standard recognition data;

and determining the confidence corresponding to each preset voice model according to the probability.

5. The method according to claim 2, wherein determining weights corresponding to the plurality of preset speech models according to the recognition result and the confidence corresponding to each preset speech model comprises:

obtaining a plurality of recognition results of the sample voice in the plurality of preset voice models, and determining a first feature vector of the sample voice according to the plurality of recognition results;

obtaining a plurality of confidence degrees of the sample voice in the preset voice models, and determining a second feature vector of the sample voice according to the confidence degrees;

and inputting the first feature vector and the second feature vector into a preset neural network model to obtain weights corresponding to the plurality of preset voice models.

6. The method according to claim 1, wherein at least one target speech model is determined from the plurality of preset speech models according to a weight corresponding to each preset speech model in the plurality of preset speech models, and before the weight of each preset speech model represents a confidence of a recognition result of the preset speech model, the method further comprises:

determining the identity information of the target object corresponding to the voice data to be processed;

and determining the calling authority of the target object according to the identity information, wherein the calling authority is used for indicating a model list which can process the voice data to be processed corresponding to the target object in a plurality of preset voice models, and different preset recognition models are used for recognizing the voice data with different structures.

7. An apparatus for processing voice data, comprising:

the acquisition module is used for acquiring voice data to be processed;

the configuration module is used for determining at least one target voice model from a plurality of preset voice models according to the weight corresponding to each preset voice model in the preset voice models, and the weight of each preset voice model represents the confidence coefficient of the recognition result of the preset voice model;

and the determining module is used for processing the voice data to be processed through the at least one target voice model.

8. The apparatus of claim 7, further comprising:

the sample module is used for acquiring sample voices for training the preset voice models; respectively processing the sample voice through the plurality of preset voice models to obtain a recognition result and a confidence degree corresponding to each preset voice model; and determining weights corresponding to the preset voice models according to the recognition results and the confidence degrees corresponding to the preset voice models.

9. A computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to carry out the method of any one of claims 1 to 6 when executed.

10. An electronic device comprising a memory and a processor, wherein the memory has stored therein a computer program, and wherein the processor is arranged to execute the computer program to perform the method of any of claims 1 to 6.

Technical Field

The present invention relates to the field of communications, and in particular, to a method and an apparatus for processing voice data, a storage medium, and an electronic apparatus.

Background

In existing voice dialog systems, natural voice audio data from a user is obtained from an input device through a voice interaction system, and the audio data is input to one or more voice recognition engines to recognize the user's voice, thereby obtaining a voice recognition result.

The identification of a single engine generally has respective problems, particularly a cloud large model, each engine has respective advantages and disadvantages, and it is generally expected that the engines can make up for each other to improve the identification effect. This involves the identification of multiple engines.

In general, a multi-engine is used in which speech data of a user is input to a plurality of engines, recognition results of all the engines are obtained, and then a certain calculation is performed to obtain a final result. However, the problem that the interactive response time of different speech recognition engines is different exists, if all the engines pass through, the user must wait for the last recognition result to arrive and then make subsequent judgment, but the mode of obtaining a better recognition result at the expense of time has a long wait time during real user interactive experience, which seriously affects the interactive experience.

In the related art, an effective technical scheme is not provided for solving the problems that the recognition time is long and the accuracy of a recognition result cannot be determined when a plurality of speech recognition engines (namely speech models) are used for speech recognition.

Disclosure of Invention

The embodiment of the invention provides a method and a device for processing voice data, a storage medium and an electronic device, which are used for at least solving the problems that in the related art, when a plurality of voice recognition engines (namely voice models) are used for voice recognition, the recognition time is long, the accuracy of a recognition result cannot be determined and the like.

According to an embodiment of the present invention, there is provided a method for processing voice data, including: acquiring voice data to be processed; determining at least one target voice model from a plurality of preset voice models according to the weight corresponding to each preset voice model in the preset voice models, wherein the weight of each preset voice model represents the confidence coefficient of the recognition result of the preset voice model; and processing the voice data to be processed through the at least one target voice model.

In an exemplary embodiment, before acquiring the voice data to be processed, the method further includes: obtaining sample voice for training the preset voice models; respectively processing the sample voice through the plurality of preset voice models to obtain a recognition result and a confidence degree corresponding to each preset voice model; and determining weights corresponding to the preset voice models according to the recognition results and the confidence degrees corresponding to the preset voice models.

In an exemplary embodiment, the processing the sample speech through the plurality of preset speech models respectively to obtain the recognition result corresponding to each preset speech model includes: obtaining standard identification data of the sample voice, wherein the standard identification data is used for indicating the sample voice to correctly analyze corresponding text content; determining the difference between the standard recognition data and the recognition data obtained by processing the sample voice by each preset voice model; and determining the recognition result of each preset voice model for the sample voice according to the difference.

In an exemplary embodiment, the processing the sample speech through the plurality of preset speech models respectively to obtain the confidence corresponding to each preset speech model includes: obtaining a confidence interval corresponding to the sample voice; determining the probability of the recognition value of each preset voice model for the sample voice processing and the confidence interval, wherein the recognition value is used for indicating the number of repeated words of the recognition data of each preset voice model for the sample voice recognition and the standard recognition data; and determining the confidence corresponding to each preset voice model according to the probability.

In an exemplary embodiment, determining weights corresponding to the preset speech models according to the recognition results and the confidence degrees corresponding to the preset speech models includes: obtaining a plurality of recognition results of the sample voice in the plurality of preset voice models, and determining a first feature vector of the sample voice according to the plurality of recognition results; obtaining a plurality of confidence degrees of the sample voice in the preset voice models, and determining a second feature vector of the sample voice according to the confidence degrees; and inputting the first feature vector and the second feature vector into a preset neural network model to obtain weights corresponding to the plurality of preset voice models.

In an exemplary embodiment, before determining at least one target speech model from the plurality of preset speech models according to a weight corresponding to each preset speech model in the plurality of preset speech models, and the weight of each preset speech model represents a confidence of a recognition result of the preset speech model, the method further includes: determining the identity information of the target object corresponding to the voice data to be processed; and determining the calling authority of the target object according to the identity information, wherein the calling authority is used for indicating a model list which can process the voice data to be processed corresponding to the target object in a plurality of preset voice models, and different preset recognition models are used for recognizing the voice data with different structures.

According to another embodiment of the present invention, there is provided a processing apparatus of voice data, including: the acquisition module is used for acquiring voice data to be processed; the configuration module is used for performing recognition configuration on the voice data according to a preset recognition model, wherein the preset recognition model is a model which is composed of a plurality of preset voice models and is used for recognizing voice, the preset recognition model comprises weights corresponding to the preset voice models, and the weights are used for indicating recognition results and weighting coefficients of confidence degrees corresponding to different preset voice models; and the determining module is used for determining at least one target voice model from the plurality of preset voice models to perform recognition processing on the voice data to be processed under the condition that the content corresponding to the recognition configuration is determined.

In an exemplary embodiment, the apparatus further includes: the sample module is used for acquiring sample voices for training the preset voice models; respectively processing the sample voice through the plurality of preset voice models to obtain a recognition result and a confidence degree corresponding to each preset voice model; and determining weights corresponding to the preset voice models according to the recognition results and the confidence degrees corresponding to the preset voice models.

In an exemplary embodiment, the sample module is further configured to obtain standard recognition data of the sample speech, where the standard recognition data is used to indicate that the sample speech correctly parses the corresponding text content; determining the difference between the standard recognition data and the recognition data obtained by processing the sample voice by each preset voice model; and determining the recognition result of each preset voice model for the sample voice according to the difference.

In an exemplary embodiment, the sample module is further configured to obtain a confidence interval corresponding to the sample voice; determining the probability of the recognition value of each preset voice model for the sample voice processing and the confidence interval, wherein the recognition value is used for indicating the number of repeated words of the recognition data of each preset voice model for the sample voice recognition and the standard recognition data; and determining the confidence corresponding to each preset voice model according to the probability.

In an exemplary embodiment, the sample module is further configured to obtain a plurality of recognition results of the sample speech in the plurality of preset speech models, and determine a first feature vector of the sample speech according to the plurality of recognition results; obtaining a plurality of confidence degrees of the sample voice in the preset voice models, and determining a second feature vector of the sample voice according to the confidence degrees; and inputting the first feature vector and the second feature vector into a preset neural network model to obtain weights corresponding to the plurality of preset voice models.

In an exemplary embodiment, the apparatus further includes: the authority module is used for determining the identity information of the target object corresponding to the voice data to be processed; and determining the calling authority of the target object according to the identity information, wherein the calling authority is used for indicating a model list which can process the voice data to be processed corresponding to the target object in a plurality of preset voice models, and different preset recognition models are used for recognizing the voice data with different structures.

According to a further embodiment of the present invention, there is also provided a storage medium having a computer program stored therein, wherein the computer program is arranged to perform the steps of any of the above method embodiments when executed.

According to yet another embodiment of the present invention, there is also provided an electronic device, including a memory in which a computer program is stored and a processor configured to execute the computer program to perform the steps in any of the above method embodiments.

By the invention, the voice data to be processed is obtained; determining at least one target voice model from a plurality of preset voice models according to the weight corresponding to each preset voice model in the plurality of preset voice models, wherein the weight of each preset voice model represents the confidence coefficient of the recognition result of the preset voice model; the voice data to be processed is processed through at least one target voice model, namely, the voice data to be processed is processed through determining the weight corresponding to each preset voice model in a plurality of preset voice models, and at least one target voice model which accords with the voice data to be processed is selected from the weights to be processed, so that more accurate voice results are fed back to a target object.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention without limiting the invention. In the drawings:

fig. 1 is a block diagram of a hardware configuration of a computer terminal of a method for processing voice data according to an embodiment of the present invention;

fig. 2 is a flowchart of a processing method of voice data according to an embodiment of the present invention;

fig. 3 is a block diagram (one) of the structure of a speech data processing apparatus according to an embodiment of the present invention;

fig. 4 is a block diagram (ii) of the configuration of a speech data processing apparatus according to an embodiment of the present invention.

Detailed Description

The invention will be described in detail hereinafter with reference to the accompanying drawings in conjunction with embodiments. It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.

It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order.

The method provided by the embodiment of the application can be executed in a computer terminal or a similar operation device of an equipment terminal. Taking the example of being operated on a computer terminal, fig. 1 is a hardware structure block diagram of a computer terminal of a method for processing voice data according to an embodiment of the present invention. As shown in fig. 1, the computer terminal may include one or more (only one shown in fig. 1) processors 102 (the processors 102 may include, but are not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, and in an exemplary embodiment, may also include a transmission device 106 for communication functions and an input-output device 108. It will be understood by those skilled in the art that the structure shown in fig. 1 is only an illustration and is not intended to limit the structure of the computer terminal. For example, the computer terminal may also include more or fewer components than shown in FIG. 1, or have a different configuration with equivalent functionality to that shown in FIG. 1 or with more functionality than that shown in FIG. 1.

The memory 104 may be used to store a computer program, for example, a software program and a module of application software, such as a computer program corresponding to the processing method of voice data in the embodiment of the present invention, and the processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, so as to implement the above-mentioned method. The memory 104 may include high speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include memory located remotely from the processor 102, which may be connected to a computer terminal over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The transmission device 106 is used for receiving or transmitting data via a network. Specific examples of the network described above may include a wireless network provided by a communication provider of the computer terminal. In one example, the transmission device 106 includes a Network adapter (NIC), which can be connected to other Network devices through a base station so as to communicate with the internet. In one example, the transmission device 106 may be a Radio Frequency (RF) module, which is used for communicating with the internet in a wireless manner.

In the present embodiment, a method for processing voice data is provided, and fig. 2 is a flowchart of a method for processing voice data according to an embodiment of the present invention, where the flowchart includes the following steps:

step S202, acquiring voice data to be processed;

step S204, determining at least one target voice model from a plurality of preset voice models according to the weight corresponding to each preset voice model in the preset voice models, wherein the weight of each preset voice model represents the confidence coefficient of the recognition result of the preset voice model;

step S206, processing the voice data to be processed through the at least one target voice model.

Through the steps, voice data to be processed are obtained; determining at least one target voice model from a plurality of preset voice models according to the weight corresponding to each preset voice model in the plurality of preset voice models, wherein the weight of each preset voice model represents the confidence coefficient of the recognition result of the preset voice model; the voice data to be processed is processed through at least one target voice model, namely, the voice data to be processed is processed through determining the weight corresponding to each preset voice model in a plurality of preset voice models, and at least one target voice model which accords with the voice data to be processed is selected from the weights to be processed, so that more accurate voice results are fed back to a target object.

It should be noted that the preset speech models have various recognition types, that is, there are a preset speech model capable of performing speech recognition, a preset speech model for performing semantic understanding, and a preset speech model for performing voiceprint recognition, which are not limited in the present invention, but similar models may be used as the preset speech models in the embodiments of the present invention.

In an exemplary embodiment, before acquiring the voice data to be processed, the method further includes: obtaining sample voice for training the preset voice models; respectively processing the sample voice through the plurality of preset voice models to obtain a recognition result and a confidence degree corresponding to each preset voice model; and determining weights corresponding to the preset voice models according to the recognition results and the confidence degrees corresponding to the preset voice models.

It should be noted that the sample voice and the voice data to be processed have the same parameter information, specifically, the parameter information may be: user ID, voiceprint characteristics, targeted voice processing devices (appliances, robots, speakers, etc.), etc.

It can be understood that, in order to ensure that the voice data can be recognized more quickly in the subsequent process, after the processing accuracy of the voice data is determined, the accuracy of different recognition models of the same semantic type is determined according to the semantic type of the corresponding content of the voice data, and then a voice data recognition list of the voice data is obtained, and when the voice data containing the same semantic type is encountered in the subsequent process, a preset recognition model corresponding to the higher recognition accuracy is selected from the voice data recognition list for recognition operation.

In an exemplary embodiment, the processing the sample speech through the plurality of preset speech models respectively to obtain the recognition result corresponding to each preset speech model includes: obtaining standard identification data of the sample voice, wherein the standard identification data is used for indicating the sample voice to correctly analyze corresponding text content; determining the difference between the standard recognition data and the recognition data obtained by processing the sample voice by each preset voice model; and determining the recognition result of each preset voice model for the sample voice according to the difference.

In an exemplary embodiment, the processing the sample speech through the plurality of preset speech models respectively to obtain the confidence corresponding to each preset speech model includes: obtaining a confidence interval corresponding to the sample voice; determining the probability of the recognition value of each preset voice model for the sample voice processing and the confidence interval, wherein the recognition value is used for indicating the number of repeated words of the recognition data of each preset voice model for the sample voice recognition and the standard recognition data; and determining the confidence corresponding to each preset voice model according to the probability.

That is, in order to ensure that the accuracy of voice data recognition is within a certain safety range, the historical word error rate corresponding to the preset recognition model is screened through the preset word error rate threshold, and then the word error rate of the preset recognition model for recognizing the voice data is ensured within the range allowed by the target object.

In an exemplary embodiment, determining weights corresponding to the preset speech models according to the recognition results and the confidence degrees corresponding to the preset speech models includes: obtaining a plurality of recognition results of the sample voice in the plurality of preset voice models, and determining a first feature vector of the sample voice according to the plurality of recognition results; obtaining a plurality of confidence degrees of the sample voice in the preset voice models, and determining a second feature vector of the sample voice according to the confidence degrees; and inputting the first feature vector and the second feature vector into a preset neural network model to obtain weights corresponding to the plurality of preset voice models.

In an exemplary embodiment, before determining at least one target speech model from the plurality of preset speech models according to a weight corresponding to each preset speech model in the plurality of preset speech models, and the weight of each preset speech model represents a confidence of a recognition result of the preset speech model, the method further includes: determining the identity information of the target object corresponding to the voice data to be processed; and determining the calling authority of the target object according to the identity information, wherein the calling authority is used for indicating a model list which can process the voice data to be processed corresponding to the target object in a plurality of preset voice models, and different preset recognition models are used for recognizing the voice data with different structures.

In short, because different target objects correspond to different identity information, the preset identification models which can be selected when the preset identification models are called are different, because the target objects can register identities on the server in advance, and allocate the calling authorities of the corresponding preset identification models to the target objects according to the registration results, namely, under the condition that the target objects are registered on the server and the identity of the target objects passes the verification, one or more preset identification models corresponding to the calling authorities can be selected from a plurality of preset identification models arranged on the server to process voice data.

In order to better understand the process of the processing method of the voice data, the following describes a flow of the processing method of the voice data with reference to two alternative embodiments.

In the intelligent voice dialogue system, in order to not influence the interactive response time, a shunting strategy for redistributing flow calls is achieved by calling a plurality of methods of a universal voice recognition engine to achieve the best user interactive experience. Because the existing multi-engine call usually identifies the same user voice data on a plurality of engines at the same time, the response time of each engine is inconsistent, and the time obtained by all results is taken as the standard, so that the longest interaction time is taken as the final response time each time, and the interaction experience of the user is seriously influenced. The advantages of multiple engines are evident, however, and can mutually offset the optimal recognition results.

In order to solve the problem, in an optional embodiment of the present invention, a method for implementing a offloading policy based on multiple speech recognition engines is mainly provided, where a policy for regularly reallocating traffic is used, each speech is recognized by only one engine, but the engine identifies the engine with the best speech among the engines, and the engine used by each user is regularly reallocated, so as to achieve the highest matching degree between the data of the user and the engines, thereby achieving the best recognition result and interactive experience, and further, by using a policy for dynamically offloading multiple engines, different engines are dynamically invoked, thereby achieving a more accurate recognition result fed back to the user within the response time invoked by a single engine, without affecting the technical effect of interactive experience.

As an alternative implementation, the solution for outputting the recognition result of the multi-pass speech recognition engine includes the following steps:

step 1, firstly, based on the existing recognition system, part of user voices enter multi-engine recognition at the same time by using man-machine conversation, and user data is screened and labeled to obtain the correct instruction requirement of a user.

Step 2, carrying out statistics on confidence coefficient values (also called confidence coefficient) obtained by the data of each engine in the step, and determining the proportion reaching the threshold value in the whole data according to the threshold value analysis of each engine;

optionally, calculation of the Confidence value: as the cloud universal model is adopted, the confidence is calculated according to different structures and results of the model.

As an alternative embodiment, the conventional model structure uses a posterior probability, that is: and determining the optimal path by using the language model and the acoustic model scoring to obtain a result of posterior probability, wherein the formula for obtaining the optimal word sequence by the voice recognition is as follows:

wherein, P (W) is the scoring of the language model, and P (X | W) is the scoring of the acoustic model.

As another alternative, the Confidence ratio may be calculated, the Confidence result of all data is calculated according to all engines, and the result is normalized by softmax,

for example, assume a total of m engines, n data:

wherein c (total) is a total confidence value, and c is_m(conf{1..n}＞thres_m) Representing whether the confidence values corresponding to the n data recognized by the M engines are larger than the preset average confidence of the M engines or not; c_MA vector of scales for indicating the confidence level of the n data in the M engine; the vectors are normalized by the softmax function: the formula is as follows:

S₁＝softmax(C_M)；

optionally, the recognition result proportion calculation: counting the recognition result of each engine according to the word error rate WER of the recognition evaluation standard, wherein the formula is as follows:

W_M＝[(1-WER₁)，...，(1-WER_m)]；

w is as described above_MA vector for identifying accuracy; also normalized by a softmax function;

S₂＝softmax(W_M)；

combining the normalized results S₁And S₂Weighted average re-measures the performance of each engine:

S＝λ₁S₁+λ₂S₂

wherein λ is₁，λ₂∈R^m，R^mFor each set of engine corresponding weight coefficients, S₁And S₂As two groups of vectors of m-dimensional characteristics, performing DNN model training by using k-fold cross validation to obtain optimal lambda₁，λ₂And thus the final allocation result S is obtained.

And 3, sequencing the S, selecting three engines with the accuracy rate of the first three, normalizing the three engines again to obtain a final weight distribution scheme, namely, the cloud end achieves the maximum improvement of the recognition rate by configuring an engine mode which can be called by a user under the condition that one engine is selected to be called by multiple engines.

And 4, periodically and repeatedly executing the steps 1-3, and automating the whole process into a mode of dynamically reallocating the engine calls according to the weight.

Alternatively, the dual engine works best in light of the actual test results (WER) of table 1 below:

TABLE 1

In summary, in the optional embodiment of the present invention, the confidence degrees and the recognition results of multiple engines are used as feature vectors, and training and tuning of weight coefficient models of different engines are performed to obtain the optimal weight result. And dynamically distributing the engines according to the weight result, so that different users can call different engines. And (3) achieving the optimal identification accuracy, regularly retraining the weight result, and dynamically distributing the engines. In addition, a multi-voice recognition engine mixed calling mode is used, the recognition accuracy is improved, a user command enters a single engine to obtain the optimal recognition result of all the engines, the response time is reduced, and further, the weights of all the engines can be automatically generated, so that different engines can be automatically called to realize a dynamic allocation strategy.

Through the above description of the embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but the former is a better implementation mode in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (e.g., ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (e.g., a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.

In this embodiment, a device for processing voice data is further provided, and the device is used to implement the foregoing embodiments and preferred embodiments, and the description of the device that has been already made is omitted. As used below, the term "module" may be a combination of software and/or hardware that implements a predetermined function. Although the means described in the embodiments below are preferably implemented in software, an implementation in hardware, or a combination of software and hardware is also possible and contemplated.

Fig. 3 is a block diagram of a configuration of a speech data processing apparatus according to an embodiment of the present invention, as shown in fig. 3, the apparatus including:

(1) an obtaining module 34, configured to obtain voice data to be processed;

(2) a configuration module 36, configured to determine at least one target speech model from the multiple preset speech models according to a weight corresponding to each preset speech model in the multiple preset speech models, where the weight of each preset speech model represents a confidence of a recognition result of the preset speech model;

(3) a determining module 38, configured to process the to-be-processed speech data through the at least one target speech model.

Acquiring voice data to be processed by the device; determining at least one target voice model from a plurality of preset voice models according to the weight corresponding to each preset voice model in the plurality of preset voice models, wherein the weight of each preset voice model represents the confidence coefficient of the recognition result of the preset voice model; the voice data to be processed is processed through at least one target voice model, namely, the voice data to be processed is processed through determining the weight corresponding to each preset voice model in a plurality of preset voice models, and at least one target voice model which accords with the voice data to be processed is selected from the weights to be processed, so that more accurate voice results are fed back to a target object.

It should be noted that the preset speech models have various recognition types, that is, there are a preset speech model capable of performing speech recognition, a preset speech model for performing semantic understanding, and a preset speech model for performing voiceprint recognition, which are not limited in the present invention, but similar models may be used as the preset speech models in the embodiments of the present invention.

Fig. 4 is a block diagram of another speech data processing apparatus according to an embodiment of the present invention, and as shown in fig. 4, the apparatus further includes: a sample module 30, a rights module 32;

in an exemplary embodiment, the apparatus further includes: the sample module is used for acquiring sample voices for training the preset voice models; respectively processing the sample voice through the plurality of preset voice models to obtain a recognition result and a confidence degree corresponding to each preset voice model; and determining weights corresponding to the preset voice models according to the recognition results and the confidence degrees corresponding to the preset voice models.

It should be noted that the sample voice and the voice data to be processed have the same parameter information, specifically, the parameter information may be: user ID, voiceprint characteristics, targeted voice processing devices (appliances, robots, speakers, etc.), etc.

It can be understood that, in order to ensure that the voice data can be recognized more quickly in the subsequent process, after the processing accuracy of the voice data is determined, the accuracy of different recognition models of the same semantic type is determined according to the semantic type of the corresponding content of the voice data, and then a voice data recognition list of the voice data is obtained, and when the voice data containing the same semantic type is encountered in the subsequent process, a preset recognition model corresponding to the higher recognition accuracy is selected from the voice data recognition list for recognition operation.

In an exemplary embodiment, the sample module is further configured to obtain standard recognition data of the sample speech, where the standard recognition data is used to indicate that the sample speech correctly parses the corresponding text content; determining the difference between the standard recognition data and the recognition data obtained by processing the sample voice by each preset voice model; and determining the recognition result of each preset voice model for the sample voice according to the difference.

In an exemplary embodiment, the sample module is further configured to obtain a confidence interval corresponding to the sample voice; determining the probability of the recognition value of each preset voice model for the sample voice processing and the confidence interval, wherein the recognition value is used for indicating the number of repeated words of the recognition data of each preset voice model for the sample voice recognition and the standard recognition data; and determining the confidence corresponding to each preset voice model according to the probability.

That is, in order to ensure that the accuracy of voice data recognition is within a certain safety range, the historical word error rate corresponding to the preset recognition model is screened through the preset word error rate threshold, and then the word error rate of the preset recognition model for recognizing the voice data is ensured within the range allowed by the target object.

In an exemplary embodiment, the sample module is further configured to obtain a plurality of recognition results of the sample speech in the plurality of preset speech models, and determine a first feature vector of the sample speech according to the plurality of recognition results; obtaining a plurality of confidence degrees of the sample voice in the preset voice models, and determining a second feature vector of the sample voice according to the confidence degrees; and inputting the first feature vector and the second feature vector into a preset neural network model to obtain weights corresponding to the plurality of preset voice models.

In an exemplary embodiment, the apparatus further includes: the authority module is used for determining the identity information of the target object corresponding to the voice data to be processed; and determining the calling authority of the target object according to the identity information, wherein the calling authority is used for indicating a model list which can process the voice data to be processed corresponding to the target object in a plurality of preset voice models, and different preset recognition models are used for recognizing the voice data with different structures.

In short, because different target objects correspond to different identity information, the preset identification models which can be selected when the preset identification models are called are different, because the target objects can register identities on the server in advance, and allocate the calling authorities of the corresponding preset identification models to the target objects according to the registration results, namely, under the condition that the target objects are registered on the server and the identity of the target objects passes the verification, one or more preset identification models corresponding to the calling authorities can be selected from a plurality of preset identification models arranged on the server to process voice data.

In the description of the present invention, it is to be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", and the like, indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience in describing the present invention and simplifying the description, but do not indicate or imply that the device or assembly referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus, should not be construed as limiting the present invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.

In the description of the present invention, it should be noted that, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, e.g., as meaning either a fixed connection, a removable connection, or an integral connection; can be mechanically or electrically connected; the two components can be directly connected or indirectly connected through an intermediate medium, and the two components can be communicated with each other. When an element is referred to as being "secured to" or "disposed on" another element, it can be directly on the other element or intervening elements may also be present. When a component is referred to as being "connected" to another element, it can be directly connected to the other element or intervening elements may also be present. The specific meaning of the above terms in the present invention can be understood in specific cases to those skilled in the art.

It should be noted that, the above modules may be implemented by software or hardware, and for the latter, the following may be implemented, but not limited to: the modules are all positioned in the same processor; alternatively, the modules are respectively located in different processors in any combination.

Embodiments of the present invention also provide a storage medium having a computer program stored therein, wherein the computer program is arranged to perform the steps of any of the above method embodiments when executed.

In an exemplary embodiment, in the present embodiment, the storage medium may be configured to store a computer program for executing the steps of:

s1, acquiring voice data to be processed;

s2, determining at least one target voice model from a plurality of preset voice models according to the weight corresponding to each preset voice model in the preset voice models, wherein the weight of each preset voice model represents the confidence coefficient of the recognition result of the preset voice model;

s3, processing the voice data to be processed through the at least one target voice model.

In an exemplary embodiment, in the present embodiment, the storage medium may include, but is not limited to: various media capable of storing computer programs, such as a usb disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic disk, or an optical disk.

Embodiments of the present invention also provide an electronic device comprising a memory having a computer program stored therein and a processor arranged to run the computer program to perform the steps of any of the above method embodiments.

In an exemplary embodiment, the electronic apparatus may further include a transmission device and an input/output device, wherein the transmission device is connected to the processor, and the input/output device is connected to the processor.

In an exemplary embodiment, in the present embodiment, the processor may be configured to execute the following steps by a computer program:

s1, acquiring voice data to be processed;

s2, determining at least one target voice model from a plurality of preset voice models according to the weight corresponding to each preset voice model in the preset voice models, wherein the weight of each preset voice model represents the confidence coefficient of the recognition result of the preset voice model;

s3, processing the voice data to be processed through the at least one target voice model.

In an exemplary embodiment, for specific examples in this embodiment, reference may be made to the examples described in the above embodiments and optional implementation manners, and details of this embodiment are not described herein again.

It will be apparent to those skilled in the art that the various modules or steps of the invention described above may be implemented using a general purpose computing device, which may be centralized on a single computing device or distributed across a network of computing devices, and in one exemplary embodiment may be implemented using program code executable by a computing device, such that the steps shown and described may be executed by a computing device stored in a memory device and, in some cases, executed in a sequence different from that shown and described herein, or separately fabricated into individual integrated circuit modules, or multiple ones of them fabricated into a single integrated circuit module. Thus, the present invention is not limited to any specific combination of hardware and software.

The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the principle of the present invention should be included in the protection scope of the present invention.