Stutter speech recognition method and device, computer equipment and storage medium

Abstract

The invention relates to the technical field of artificial intelligence, and provides a stutter speech recognition method and device, computer equipment and a storage medium, and the method comprises the steps: obtaining initial text information used for recognizing a target stutter speech; inputting the initial text information into a target stutter predictor for prediction, and generating target text information according to a prediction result of the target stutter predictor; and inputting the target text information to a preset speech recognition model, and recognizing the target stutter speech corresponding to the target text information in the preset speech recognition model, thereby improving the accuracy of stutter speech recognition.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a stuttering speech recognition method, apparatus, computer equipment and storage medium. Background Art

[0002] In the field of artificial intelligence technology, data imbalance is a common problem, especially in the specific application of speech recognition (Automatic Speech Recognition, ASR) and text-to-speech (Text-to-Speech, TTS) technology. Due to the scarcity of stuttering data sets, the existing preset ASR model has insufficient recognition accuracy for stuttering speech, resulting in low recognition accuracy for stuttering speech. Therefore, how to predict stuttering events and improve the accuracy of stuttering speech recognition has become a technical problem that needs to be solved urgently. Summary of the invention

[0003] The main purpose of this application is to provide a stuttering speech recognition method, apparatus, computer equipment and storage medium to improve the accuracy of stuttering speech recognition.

[0004] In a first aspect, the present application provides a stuttering speech recognition method, the stuttering speech recognition method comprising the following steps:

[0005] Acquire initial text information for identifying a target stuttering speech; input the initial text information into a target stuttering predictor for prediction, and generate target text information according to the prediction result of the target stuttering predictor; input the target text information into a preset speech recognition model, and identify the target stuttering speech corresponding to the target text information in the preset speech recognition model.

[0006] In a second aspect, the present application further provides a stuttering speech recognition device, the stuttering speech recognition device comprising:

[0007] An initial text information determination module is used to obtain initial text information for identifying a target stuttering speech; a prediction module is used to input the initial text information into a target stuttering predictor for prediction, and generate target text information according to the prediction result of the target stuttering predictor; a target speech recognition module is used to input the target text information into a preset speech recognition model, and identify the target stuttering speech corresponding to the target text information in the preset speech recognition model.

[0008] In a third aspect, the present application also provides a computer device, comprising a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein when the computer program is executed by the processor, the stuttering speech recognition method as described above is implemented.

[0009] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the stuttering speech recognition method as described above is implemented.

[0010] The present application provides a stuttering speech recognition method, apparatus, computer equipment and storage medium, which obtains initial text information for identifying a target stuttering speech; inputs the initial text information into a target stuttering predictor for prediction, and generates target text information according to the prediction result of the target stuttering predictor; inputs the target text information into a preset speech recognition model, and recognizes the target stuttering speech corresponding to the target text information in the preset speech recognition model, inputs the initial text information into the target stuttering predictor, and obtains a target prediction category according to the prediction result of the target stuttering predictor, adds the target prediction category to a preset position of the initial text information, obtains the target text information, and inputs the target stuttering speech into the preset speech recognition model to obtain the target stuttering speech, thereby improving the accuracy of stuttering speech recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0012] Figure 1 A flowchart of a stuttering speech recognition method provided in one embodiment of the present application.

[0013] Figure 2 A schematic flowchart of determining target text information provided by an embodiment of the present application.

[0014] Figure 3 A schematic diagram of a preset speech recognition model provided in an embodiment of the present application.

[0015] Figure 4 Another schematic flowchart of determining target text information is provided in an embodiment of the present application.

[0016] Figure 5 A schematic diagram of a stuttering speech recognition device provided in one embodiment of the present application.

[0017] Figure 6 A schematic block diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0019] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0020] The embodiments of the present application provide a stuttering speech recognition method, apparatus, computer equipment and storage medium.

[0021] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0022] Please refer to Figure 1 , Figure 1 A flowchart of a stuttering speech recognition method provided for one embodiment of the present application. The stuttering speech recognition method can be used in a terminal or a server to improve the recognition accuracy of stuttering speech. The terminal can be an electronic device such as a mobile phone, a tablet computer, a laptop computer, a desktop computer, a personal digital assistant, and a wearable device; the server can be an independent server, or a server cluster, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms. Figure 1 As shown, the stuttering speech recognition method specifically includes steps S101 to S103.

[0023] S101: Acquire initial text information for recognizing a target stuttered speech.

[0024] The number of initial text information is at least one.

[0025] Specifically, the types of the acquired initial text information may be four initial text types, namely, initial text information with sound extension, initial text information with pause type, initial text information with both sound extension and pause, and normal text information (non-pause type).

[0026] It should be noted that in actual applications, other suitable text categories may be selected, or some categories may be added or deleted according to the needs of the actual application, and this application does not impose any restrictions on this.

[0027] For example, a text message with extended sound may be: Today, the weather is very good. Let's go to the park to play. A text message with a pause may be: Today, the weather is very good. Let's go to the park to play. A text message with both extended sound and pause may be: Today, the weather is very good. Let's go to the park to play.

[0028] It should be noted that the different types of initial text information obtained above are only for illustrative purposes. In practical applications, any other appropriate initial text information may be obtained according to actual needs, and this application does not impose any limitation on this.

[0029] S102: Input the initial text information into a target stuttering predictor for prediction, and generate target text information according to the prediction result of the target stuttering predictor.

[0030] Specifically, before the initial text information is input into the target stuttering predictor, the text format of the initial text information is modified according to the text type acceptable to the target stuttering predictor to ensure that the target stuttering predictor can be compatible with receiving the initial text information. The text type acceptable to the target stuttering predictor can be any one of: .txt file, .doc, and .docx.

[0031] It should be noted that the text types acceptable to the above-mentioned target stuttering predictor are only for illustrative purposes. In practical applications, any other suitable text type may be selected according to specific circumstances, and this application does not impose any limitation on this.

[0032] Exemplarily, the initial text information can be input into the target stuttering predictor based on the application programming interface (API) of the target stuttering predictor, wherein the API of the target stuttering predictor can be any suitable API type such as a Web API, a library API, an operating system API, a database API, etc., and this application does not limit this.

[0033] Specifically, after the initial text information is input into a target stuttering predictor for prediction, target text information is generated according to the prediction result of the target stuttering predictor.

[0034] In some embodiments, all initial text information includes first text information and second text information. Before adding the target prediction category to a preset position of the initial text information to generate the target text information, the method also includes: classifying all the initial text information based on a preset selector to obtain the first text information and the second text information classified by the preset selector; adding the target prediction category corresponding to the first text information to the preset position of the first text information to generate the target text information corresponding to the first text information.

[0035] The number of the first text information and the second text information is not greater than the number of the initial text information.

[0036] Exemplarily, the preset selector may be a decision tree, a support vector machine, a naive Bayes classifier, or the like.

[0037] It should be noted that the preset selectors listed above are only for illustrative purposes. In practical applications, other suitable preset selectors may be selected according to actual needs, and the present application does not impose any limitation on this.

[0038] After the initial text information is classified according to the preset selector to obtain the first text information and the second text information, the target prediction category corresponding to each first text information is obtained, and the target prediction category corresponding to each first text information is added to the preset position of the corresponding first text information to obtain the target text information corresponding to the first text information. The obtained second text information is not subjected to the relevant processing of adding the target prediction category.

[0039] The above-mentioned processing method, which only determines the target text information corresponding to the first text information based on the preset selector but does not perform any operation on the second text information, realizes the controllability of the amount of target text information obtained, and can generate a specific amount of target text information according to actual needs, thereby avoiding the generation of too many target text information and wasting resources for generating target text information, and improving the efficiency of generating target text information.

[0040] It should be noted that the preset position can be any suitable position in the first text information, such as the front end of the first text information, the middle of the first text information, or the end of the first text information. The specific position for adding the target prediction category in the first text information can be reasonably set according to actual conditions, and this application does not impose any restrictions on this.

[0041] In some embodiments, the classification processing of all the initial text information based on the preset selector to obtain the first text information and the second text information classified by the preset selector includes: determining a first parameter; if the first parameter is greater than or equal to the number of all the initial text information, the preset selector determines all the initial text information as the first text information; if the first parameter is less than the number of all the initial text information, the preset selector determines the initial text information from the first initial text information to the initial text information at the position of the first parameter as the first text information, and determines the remaining initial text information as the second text information.

[0042] Among them, the preset selector can also be any suitable selector such as a keyword selector, a regular expression selector, a sentence selector, a topic selector, etc., and this application does not impose any limitation on this.

[0043] Exemplarily, if the preset selector is a keyword selector, the initial text information is input into the keyword selector, and after the selection of the keyword selector, the above initial text information can be determined as any one of a text information with repeated words, a text information with extended sound, or a text information with a pause. For example, the initial text information is: I want to eat apples. The initial text information is then input into the keyword selector, and after screening by the keyword selector, the initial text information is classified into a category of text information with extended sound. Similarly, each other initial text information is input into the keyword selector in turn, and each initial text information will be classified into a corresponding text type by the keyword selector. The specific classification process and the above-mentioned types of each initial text information by the keyword selector will not be repeated in this application.

[0044] Exemplarily, the first parameter may be represented by the letter N, and the specific value of the first parameter N may be set to any suitable positive integer such as 10, 20, 100, etc., and the present application does not impose any limitation on this.

[0045] It should be noted that the above-mentioned first parameter can also be represented by letters such as T, X, etc. In practical applications, any other suitable letters can be selected as needed, and this application does not impose any limitation on this.

[0046] Exemplarily, if the number of all the initial text information is 20, N=30, and the first parameter is greater than the number of all the initial text information, then the preset selector determines all the initial text information as the first text information. Then, the target prediction category corresponding to each of the first text information is added to the preset position of the first text information, and the target text information corresponding to the first text information is generated. Similarly, when the first parameter is equal to the number of all the initial text information, all the initial text information is also determined as the first text information, and the target text information corresponding to all the first text information is obtained.

[0047] For example, if the total number of the initial text messages is 20, N=15, then

[0048] If the first parameter is less than the number of all the initial text information, the preset selector determines the first initial text information to the initial text information at the first parameter position as the first text information, and determines the remaining initial text information as the second text information. For example, all the initial text information are: initial text information 1, initial text information 2, initial text information 3, ..., initial text information 20, then the first initial text information 1 to the initial text information 15 at the first parameter position are all determined as the first text information, that is, the first 15 initial text information, that is, initial text information 1, initial text information 2, initial text information 3, ..., initial text information 15, are determined as the first text information. The remaining initial text information, that is, initial text information 16, initial text information 17, ..., initial text information 20, are all determined as the second text information.

[0049] In another example, if the categories corresponding to the first five target text information are all text information with repeated words, and the subsequent ones are target text information corresponding to the other three categories of initial text information, and only the target stuttering speech corresponding to the first five target text information with repeated words is to be analyzed, the first parameter N can be set to 5. Similarly, in other cases, the first parameter can also be set to other appropriate values, and this application does not limit this.

[0050] In summary, the first N initial text information is selected from the starting position according to the set first parameter N, and the first N initial text information is determined as the first text information, the target text information corresponding to the first text information is determined, and the remaining initial text information is determined as the second text information. When the number of initial text information is less than N, all the initial text information is directly determined as the first text information.

[0051] It should be noted that the purpose of setting the first parameter is to achieve controllability of the amount of target text information.

[0052] See also Figure 3 , Figure 3 The present invention is a schematic diagram of a preset speech recognition model in an embodiment of the present application. In some embodiments, the preset speech recognition model includes a front-end model and an acoustic model, and the step of inputting the target text information into the preset speech recognition model and identifying the target stuttering speech corresponding to the target text information in the preset speech recognition model includes: inputting the target text information into the front-end model to obtain the target speech information to be generated output by the front-end model; inputting the target speech information to be generated into the acoustic model, performing speech recognition in the acoustic model, and obtaining the target stuttering speech.

[0053] The target voice information to be generated is voice information containing phonemes.

[0054] For example, "It's very hot today, and the sky is clear and cloudless" can be processed by the front-end model to become "Uh, it's very hot today, and the sky is clear and cloudless", and then converted into the corresponding target speech information to be generated in the front-end model, and the target speech information to be generated is input into the acoustic model for speech recognition to obtain the corresponding target stuttering speech. After the joint training of the above front-end model and the acoustic model, the target stuttering speech can be finally obtained.

[0055] Among them, the above-mentioned front-end model can be any one of a probabilistic language model, a rule-based language model, a speech rate predictor, etc., and the above-mentioned acoustic model can be any one of fastspeech, tracotron, and WaveNet.

[0056] It should be noted that the specific types of the above-mentioned front-end models and acoustic models are only used for illustrative purposes. In practical applications, any other suitable model may be selected, and this application does not impose any restrictions on this.

[0057] In some embodiments, before inputting the initial text information into the target stuttering predictor for prediction, the method further includes: training a prediction module in the target stuttering predictor based on a preset loss formula to determine a prediction weight of the preset loss formula.

[0058] Exemplarily, the preset loss formula is:

[0059]

[0060] Wherein, L represents the loss value, y1 represents the first prediction category, y2 represents the second prediction category, σ represents the weight of the second prediction category, wherein the first prediction category is a text with pauses, and the second prediction category is a non-pause text, represents the first predicted probability, represents the second predicted probability.

[0061] Exemplarily, if the second prediction category is small, in order to balance the loss of the first prediction category and the second prediction category and prevent the stuttering predictor from being biased towards the first prediction category, the penalty for prediction errors of the second prediction category may be increased by increasing the value of σ.

[0062] As another example, in this solution, the number of initial text information with pause and extension labels is small, which is a sparse label. The above loss formula can solve the problem of processing the imbalance of initial text information to achieve the penalty of prediction errors of the target text information corresponding to the initial text information with pause and extension labels.

[0063] It should be noted that the above preset loss formula only considers the two situations of pause and non-pause. In practical applications, any combination of two situations can be selected according to actual needs, and this application does not limit this.

[0064] S103: Input the target text information into a preset speech recognition model, and recognize the target stuttering speech corresponding to the target text information in the preset speech recognition model.

[0065] Specifically, determine the interface type or text type that the preset speech recognition model can receive the target text information, and convert the target text information into a corresponding text type. For example, the interface type of the preset speech recognition model can be a command line interface, an application programming interface, a Web service, etc. The target text information is input into the preset speech recognition model based on the interface or API corresponding to the preset speech recognition model.

[0066] In some embodiments, after inputting the target text information into a preset speech recognition model to obtain a target stuttering speech corresponding to the target text information, the method further includes: inputting the target stuttering speech into a preset speech recognition model, training the preset speech recognition model, and obtaining a target speech recognition model corresponding to the preset speech recognition model.

[0067] Specifically, the obtained multiple target stuttering speech are sequentially input into a preset speech recognition model, and the preset speech recognition model is trained based on the target stuttering speech to improve the recognition accuracy of the preset speech recognition model.

[0068] Exemplarily, the preset speech recognition model may be any one of Fastspeech2 and Tactron.

[0069] It should be noted that Fastspeech2 and Tactron given in the above-mentioned preset speech recognition models are only used as examples. In actual applications, any other suitable preset speech recognition model can be selected according to the specific situation, and this application does not impose any restrictions on this.

[0070] In the stuttering speech recognition method provided in the embodiment of the present application, initial text information for identifying the target stuttering speech is obtained; the initial text information is input into the target stuttering predictor for prediction, and the target text information is generated according to the prediction result of the target stuttering predictor; the target text information is input into the preset speech recognition model, and the target stuttering speech corresponding to the target text information is recognized in the preset speech recognition model. By inputting the obtained target text information into the preset loss formula, the prediction error of the target text information corresponding to the initial text information with pause and extension tags is punished, and the accuracy of the prediction of the initial text information with pause and extension tags is improved; before obtaining the target text information, by inputting the initial text information into the preset selector, multiple initial text information will be classified, and the target text information corresponding to the first text information after the initial text information is classified is obtained, so as to realize the controllability of the number of generated target text information and improve the efficiency of generating target text information.

[0071] See also Figure 2 , Figure 2 A schematic flow chart of determining target text information provided by an embodiment of the present application. Figure 2 As shown, the stuttering speech recognition method specifically includes steps S201 to S205.

[0072] S201: Determine a first prediction category, a second prediction category, a third prediction category, and a fourth prediction category.

[0073] Specifically, the first prediction category is the initial text information corresponding to the pause category; the second prediction category is the initial text information corresponding to the non-pause category; the third prediction category is the initial text information with sound extension; and the fourth prediction category is the initial text information with both sound extension and pause.

[0074] S202: Input the initial text information into the target stuttering predictor to obtain a first prediction probability corresponding to the first prediction category, a second prediction probability corresponding to the second prediction category, a third prediction probability corresponding to the third prediction category, and a fourth prediction probability corresponding to the fourth prediction category.

[0075] Specifically, each time an initial text information is input into the target stuttering predictor, the target stuttering predictor predicts the specific values ​​of the first prediction probability corresponding to the first prediction category, the second prediction probability corresponding to the second prediction category, the third prediction probability corresponding to the third prediction category, and the fourth prediction probability corresponding to the fourth prediction category for the initial text information.

[0076] For example, if the initial text information is: I want to eat ~~ apples. After the initial text information is input into the target stuttering predictor, the target stuttering predictor predicts the initial text information, and the first prediction probability corresponding to the first prediction category (that is, the initial text information corresponding to the pause category) is 30%; the second prediction probability corresponding to the second prediction category (that is, the initial text information corresponding to the non-pause category) is 50%; the third prediction probability corresponding to the third prediction category (that is, the initial text information with sound extension) is 90%; the fourth prediction probability corresponding to the fourth prediction category (that is, the initial text information with both sound extension and pause) is 40%. The specific steps of inputting other initial text information into the target stuttering predictor for prediction are similar to the steps given in the above embodiment, and the present application will not repeat them.

[0077] It should be noted that the initial text information in the above embodiment and the specific values ​​of the first prediction probability, the second prediction probability, the third prediction probability and the fourth prediction probability corresponding to the initial text information are only used for illustrative purposes. In actual applications, the corresponding specific values ​​of the first prediction probability, the second prediction probability, the third prediction probability and the fourth prediction probability can be obtained according to the specific application situation and the specific prediction situation of the target stuttering predictor. This application does not impose any restrictions on this.

[0078] S203: Taking the maximum value among the first prediction probability, the second prediction probability, the third prediction probability, and the fourth prediction probability as the target prediction probability.

[0079] Specifically, after obtaining the first prediction probability, the second prediction probability, the third prediction probability and the fourth prediction probability corresponding to the initial text information, the maximum value among the first prediction probability, the second prediction probability, the third prediction probability and the fourth prediction probability is taken as the target prediction probability.

[0080] For example, if the first prediction probability corresponding to a certain initial text information is 20%, the second prediction probability is 60%, the third prediction probability is 30%, and the fourth prediction probability is 40%, the second prediction probability is used as the target prediction probability. Similarly, the target prediction probabilities corresponding to other initial text information can be obtained, which will not be described in detail in this application.

[0081] S204: Using the prediction category corresponding to the target prediction probability as the target prediction category corresponding to the initial text information.

[0082] Exemplarily, if the second prediction probability has the largest value among the first prediction probability, the second prediction probability, the third prediction probability, and the fourth prediction probability, then the second prediction probability is the target prediction probability. The prediction category corresponding to the second prediction probability is the initial text information corresponding to the non-pause category, then the target prediction category corresponding to the initial text information is the non-pause category. The target prediction category corresponding to other initial text information can be obtained in a similar manner, and this application will not go into details.

[0083] S205: Add the target prediction category to a preset position of the initial text information to generate the target text information.

[0084] The preset position may be any suitable position such as the front end of the initial text information, the middle of the initial text information, or the back end of the initial text information, and this application will not elaborate on this.

[0085] Exemplarily, after determining the target prediction category corresponding to the initial text information, the target prediction category is added to the preset position of the initial text information to generate the target text information. For example, the initial text information is: I want to...eat corn. After being predicted by the target stuttering predictor, the target prediction category obtained is the first prediction category (that is, the initial text information corresponding to the pause category), then the target prediction category is added to the front end of the initial text information (assuming the preset position is the front end), and the target text information is obtained as: [pause category] I want to...eat corn. For example, the initial text information is: I want to~eat corn. After being predicted by the target stuttering predictor, the target prediction category obtained is the third prediction category (that is, the initial text information with sound extension), then the target prediction category is added to the front end of the initial text information (assuming the preset position is the front end), and the target text information is obtained as: [sound extension category] I want to~eat corn. Similarly, when the target prediction category is the third prediction category and the fourth prediction category respectively, the third prediction category or the fourth prediction category is added to the appropriate preset position of the initial text information to obtain the corresponding target text information, which is not repeated in this application.

[0086] It should be noted that the above process of determining the target text information can also be referred to Figure 4 , Figure 4 Another schematic flow chart for determining target text information provided by an embodiment of the present application is as follows: Figure 4 The process or steps in Figure 2 The specific process is similar to that in the present application and will not be repeated here.

[0087] The method for determining target text information provided by the embodiment of the present application includes: determining a first prediction category, a second prediction category, a third prediction category and a fourth prediction category; inputting the initial text information into the target stuttering predictor to obtain a first prediction probability corresponding to the first prediction category, a second prediction probability corresponding to the second prediction category, a third prediction probability corresponding to the third prediction category and a fourth prediction probability corresponding to the fourth prediction category; taking the maximum value among the first prediction probability, the second prediction probability, the third prediction probability and the fourth prediction probability as the target prediction probability; taking the prediction category corresponding to the target prediction probability as the target prediction category corresponding to the initial text information; adding the target prediction category to a preset position of the initial text information to generate the target text information, determining the first prediction category, the second prediction category, the third prediction category and the fourth prediction category, and calculating the first prediction probability, the second prediction probability, the third prediction probability and the fourth prediction probability in the target stuttering predictor, and determining the specific values ​​of the first prediction probability, the second prediction probability, the third prediction probability and the fourth prediction probability in a specific calculation manner, rather than relying solely on subjective feelings, so as to avoid being affected by other external factors and affecting the accuracy of the data, thereby improving the accuracy of the prediction. The maximum value among the first prediction probability, the second prediction probability, the third prediction probability and the fourth prediction probability is taken as the target prediction probability, and the target prediction category corresponding to the target prediction probability is determined, and the target prediction category is added to the preset position of the initial text information to obtain the target text information. The target text information is clearly marked through the prediction of the target stuttering predictor, thereby providing an accurate text basis for the subsequent accurate generation of stuttering speech.

[0088] See also Figure 5 , Figure 5 A schematic diagram of a stuttering speech recognition device provided in an embodiment of the present application. The stuttering speech recognition device can be configured in a server or a terminal to execute the aforementioned stuttering speech recognition method.

[0089] like Figure 5 As shown, the stuttering speech recognition device 500 includes: an initial text information determination module 501, a prediction module 502, and a target speech recognition module 503.

[0090] The initial text information determination module 501 is used to obtain initial text information for identifying a target stuttered speech.

[0091] The prediction module 502 is used to input the initial text information into a target stuttering predictor for prediction, and generate target text information according to the prediction result of the target stuttering predictor.

[0092] The target speech recognition module 503 is used to input the target text information into a preset speech recognition model, and recognize the target stuttering speech corresponding to the target text information in the preset speech recognition model.

[0093] In some embodiments, the initial text information determination module 501 is specifically configured to:

[0094] Obtain initial text information for recognizing the target stuttered speech.

[0095] In some embodiments, the prediction module 502 is specifically configured to:

[0096] Determine a first prediction category, a second prediction category, a third prediction category and a fourth prediction category; input the initial text information into the target stuttering predictor, and obtain a first prediction probability corresponding to the first prediction category, a second prediction probability corresponding to the second prediction category, a third prediction probability corresponding to the third prediction category, and a fourth prediction probability corresponding to the fourth prediction category; use the maximum value among the first prediction probability, the second prediction probability, the third prediction probability and the fourth prediction probability as the target prediction probability; use the prediction category corresponding to the target prediction probability as the target prediction category corresponding to the initial text information; add the target prediction category to a preset position of the initial text information to generate the target text information.

[0097] In some embodiments, the prediction module 502 is further configured to:

[0098] Classifying all the initial text information based on a preset selector to obtain the first text information and the second text information classified by the preset selector;

[0099] The target prediction category corresponding to the first text information is added to a preset position of the first text information to generate the target text information corresponding to the first text information.

[0100] In some embodiments, the prediction module 502 is further configured to:

[0101] Determine a first parameter; if the first parameter is greater than or equal to the number of all the initial text information, the preset selector determines all the initial text information as the first text information; if the first parameter is less than the number of all the initial text information, the preset selector determines the initial text information from the first initial text information to the first parameter position as the first text information, and determines the remaining initial text information as the second text information.

[0102] In some embodiments, the prediction module 502 is further configured to:

[0103] Based on a preset loss formula, a prediction module in the target stuttering predictor is trained to determine a prediction weight of the preset loss formula.

[0104] In some embodiments, the target speech recognition module 503 is specifically configured to:

[0105] The target text information is input into the front-end model to obtain the target speech information to be generated output by the front-end model; the target speech information to be generated is input into the acoustic model, speech recognition is performed in the acoustic model to obtain the target stuttering speech.

[0106] In some embodiments, the target speech recognition module 503 is further configured to:

[0107] The target stuttering speech is input into a preset speech recognition model, and the preset speech recognition model is trained to obtain a target speech recognition model corresponding to the preset speech recognition model.

[0108] It should be noted that those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described device and each module can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0109] The above-mentioned device can be implemented in the form of a computer program. Figure 6 Runs on the computer device shown.

[0110] See also Figure 6 , Figure 6 1 is a schematic block diagram of a computer device provided in an embodiment of the present application. The computer device may be a server.

[0111] See also Figure 6 The computer device includes a processor, a memory and a network interface connected through a system bus, wherein the memory may include a non-volatile storage medium and an internal memory.

[0112] The non-volatile storage medium can store an operating system and a computer program. The computer program includes program instructions, and when the program instructions are executed, the processor can execute any stuttering speech recognition method.

[0113] The processor is used to provide computing and control capabilities and support the operation of the entire computer equipment.

[0114] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor, the processor can execute any stuttering speech recognition method.

[0115] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art will appreciate that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0116] It should be understood that the processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0117] In one embodiment, the processor is used to run a computer program stored in the memory to implement the following steps:

[0118] Acquire initial text information for identifying a target stuttering speech; input the initial text information into a target stuttering predictor for prediction, and generate target text information according to the prediction result of the target stuttering predictor; input the target text information into a preset speech recognition model, and identify the target stuttering speech corresponding to the target text information in the preset speech recognition model.

[0119] In one embodiment, the inputting of the initial text information into a target stuttering predictor for prediction and generating target text information according to the prediction result of the target stuttering predictor is used to implement:

[0120] Determine a first prediction category, a second prediction category, a third prediction category and a fourth prediction category; input the initial text information into the target stuttering predictor, and obtain a first prediction probability corresponding to the first prediction category, a second prediction probability corresponding to the second prediction category, a third prediction probability corresponding to the third prediction category, and a fourth prediction probability corresponding to the fourth prediction category; use the maximum value among the first prediction probability, the second prediction probability, the third prediction probability and the fourth prediction probability as the target prediction probability; use the prediction category corresponding to the target prediction probability as the target prediction category corresponding to the initial text information; add the target prediction category to a preset position of the initial text information to generate the target text information.

[0121] In one embodiment, all the initial text information includes the first text information and the second text information, and before adding the target prediction category to the preset position of the initial text information and generating the target text information, it is used to achieve:

[0122] Based on the preset selector, all the initial text information is classified and processed to obtain the first text information and the second text information classified by the preset selector; the target prediction category corresponding to the first text information is added to the preset position of the first text information to generate the target text information corresponding to the first text information.

[0123] In one embodiment, the classification process of all the initial text information based on the preset selector to obtain the first text information and the second text information classified by the preset selector is used to achieve:

[0124] Determine a first parameter; if the first parameter is greater than or equal to the number of all the initial text information, the preset selector determines all the initial text information as the first text information; if the first parameter is less than the number of all the initial text information, the preset selector determines the initial text information from the first initial text information to the first parameter position as the first text information, and determines the remaining initial text information as the second text information.

[0125] In one embodiment, the preset speech recognition model includes a front-end model and an acoustic model, and the target text information is input into the preset speech recognition model, and the target stuttering speech corresponding to the target text information is recognized in the preset speech recognition model, so as to achieve:

[0126] The target text information is input into the front-end model to obtain the target speech information to be generated output by the front-end model; the target speech information to be generated is input into the acoustic model, speech recognition is performed in the acoustic model to obtain the target stuttering speech.

[0127] In one embodiment, before the initial text information is input into the target stuttering predictor for prediction, it is used to implement:

[0128] Based on a preset loss formula, a prediction module in the target stuttering predictor is trained to determine a prediction weight of the preset loss formula.

[0129] In one embodiment, the inputting of the target text information into a preset speech recognition model, after the preset speech recognition model recognizes the target stuttering speech corresponding to the target text information, is used to implement:

[0130] The target stuttering speech is input into a preset speech recognition model, and the preset speech recognition model is trained to obtain a target speech recognition model corresponding to the preset speech recognition model.

[0131] A computer-readable storage medium is also provided in an embodiment of the present application. The computer-readable storage medium stores a computer program. The computer program includes program instructions. The processor executes the program instructions to implement any stuttering speech recognition method provided in the embodiments of the present application.

[0132] The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiment, such as a hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a smart memory card (Smart Medi a Card, SMC), a secure digital (Secure Digital, SD) card, a flash memory card (Flash Card), etc., equipped on the computer device.

[0133] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A stuttering speech recognition method, characterized in that: include: Acquiring initial text information for recognizing target stuttered speech; Inputting the initial text information into a target stuttering predictor for prediction, and generating target text information according to the prediction result of the target stuttering predictor; The target text information is input into a preset speech recognition model, and the target stuttering speech corresponding to the target text information is recognized in the preset speech recognition model.

2. The stuttering speech recognition method according to claim 1, characterized in that: The step of inputting the initial text information into a target stuttering predictor for prediction, and generating target text information according to the prediction result of the target stuttering predictor, comprises: determining a first prediction category, a second prediction category, a third prediction category, and a fourth prediction category; Inputting the initial text information into the target stuttering predictor, respectively obtaining a first prediction probability corresponding to the first prediction category, a second prediction probability corresponding to the second prediction category, a third prediction probability corresponding to the third prediction category, and a fourth prediction probability corresponding to the fourth prediction category; taking the maximum value among the first predicted probability, the second predicted probability, the third predicted probability, and the fourth predicted probability as the target predicted probability; Using the prediction category corresponding to the target prediction probability as the target prediction category corresponding to the initial text information; The target prediction category is added to a preset position of the initial text information to generate the target text information.

3. The stuttering speech recognition method according to claim 2, characterized in that: All the initial text information includes the first text information and the second text information. Before adding the target prediction category to a preset position of the initial text information to generate the target text information, the method further includes: Classifying all the initial text information based on a preset selector to obtain the first text information and the second text information classified by the preset selector; The target prediction category corresponding to the first text information is added to a preset position of the first text information to generate the target text information corresponding to the first text information.

4. The stuttering speech recognition method according to claim 3, characterized in that: The classifying process of all the initial text information based on the preset selector to obtain the first text information and the second text information classified by the preset selector includes: determining a first parameter; If the first parameter is greater than or equal to the number of all the initial text information, the preset selector determines all the initial text information as the first text information; If the first parameter is less than the total number of the initial text information, the preset selector determines the initial text information from the first initial text information to the initial text information at the first parameter position as the first text information, and determines the remaining initial text information as the second text information.

5. The stuttering speech recognition method according to claim 1, characterized in that: The preset speech recognition model includes a front-end model and an acoustic model, and the step of inputting the target text information into the preset speech recognition model and identifying the target stuttering speech corresponding to the target text information in the preset speech recognition model includes: Inputting the target text information into the front-end model to obtain the target speech information to be generated output by the front-end model; The target speech information to be generated is input into the acoustic model, speech recognition is performed in the acoustic model, and the target stuttering speech is obtained.

6. The stuttering speech recognition method according to claim 1, characterized in that: Before inputting the initial text information into the target stuttering predictor for prediction, the method further includes: Based on a preset loss formula, a prediction module in the target stuttering predictor is trained to determine a prediction weight of the preset loss formula.

7. The stuttering speech recognition method according to claim 1, characterized in that: After inputting the target text information into a preset speech recognition model and recognizing the target stuttering speech corresponding to the target text information in the preset speech recognition model, the method further includes: The target stuttering speech is input into a preset speech recognition model, and the preset speech recognition model is trained to obtain a target speech recognition model corresponding to the preset speech recognition model.

8. A stuttering speech recognition device, characterized in that: include: An initial text information determination module, used to obtain initial text information for identifying a target stuttered speech; A prediction module, used for inputting the initial text information into a target stuttering predictor for prediction, and generating target text information according to the prediction result of the target stuttering predictor; The target speech recognition module is used to input the target text information into a preset speech recognition model, and recognize the target stuttering speech corresponding to the target text information in the preset speech recognition model.

9. A computer device, characterized in that: The computer device includes a memory and a processor; The memory is used to store computer programs; The processor is used to execute the computer program and implement the stuttering speech recognition method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements the stuttering speech recognition method according to any one of claims 1 to 7.

Images