CN114550691A - A polyphonic word disambiguation method, device, electronic device and readable storage medium - Google Patents

Abstract

The invention relates to a method, a device, electronic equipment and a readable storage medium for disambiguating polyphone, which relate to the technical field of voice processing and comprise the following steps: dividing a text to be processed into a plurality of characters, wherein the plurality of characters comprise target polyphone characters and non-target polyphone characters; aiming at each character, acquiring a first identifier corresponding to the character; inputting the characters and the first identifications corresponding to the characters into a pre-generated target polyphonic character disambiguation model, and determining the pronunciation of the target polyphonic character according to the output of the target polyphonic character disambiguation model. The method and the device are applied to scenes for realizing the multi-tone word disambiguation in a voice synthesis system, and the target multi-tone word disambiguation model is utilized to perform the multi-tone word disambiguation on the target multi-tone word in the text to be processed, so that the prediction speed of the multi-tone word disambiguation under the scenes is improved, and further, the effect of the multi-tone word disambiguation is improved.

Description

A polyphonic word disambiguation method, device, electronic device and readable storage medium

technical field

The present application relates to the technical field of speech processing, and in particular, to a polyphonic word disambiguation method, apparatus, electronic device, and readable storage medium.

Background technique

Text To Speech (TTS) is a technology that uses a computer to convert any text into speech. For the input text, it needs to be converted into the corresponding pronunciation. Among them, the correctness of the polyphonic conversion greatly affects the user's understanding of the synthesized speech. If the polyphonic conversion is wrong, the effect of speech synthesis will be greatly reduced. . Therefore, polyphone disambiguation is an important task in speech synthesis system.

At present, the realization methods of polyphonic word disambiguation are based on decision tree, based on maximum entropy algorithm, and based on expert knowledge (a large number of rules). It is possible to pronounce a final probability value. Since this method requires preset problems and probability values, when the context or scene changes, these problems will not adapt, making the polyphonic word disambiguation effect of this method poor; The maximum entropy model is a classification model designed based on the principle of maximum entropy, which requires a large amount of data. However, if the sample data is large, it will lead to a large amount of calculation, which makes this method have certain application limitations; based on expert knowledge ( The method with a large number of rules) has the problems of laborious maintenance and easy conflict or mutual influence between the rules; therefore, the method for realizing polyphonic word disambiguation in the prior art has the problem of poor polyphonic word disambiguation effect. Therefore, there is an urgent need for a polyphonic word disambiguation method to solve the problem that the polyphonic word disambiguation effect is poor in the prior art.

SUMMARY OF THE INVENTION

In order to overcome the problems existing in the related art, the present application provides a polyphonic word disambiguation method, device, electronic device and readable storage medium.

According to a first aspect of the embodiments of the present application, a method for disambiguating a polyphonic word is provided, the method comprising:

Dividing the text to be processed into several characters, wherein the several characters include target polyphonic characters and non-target polyphonic characters;

For each character, obtain the first identifier corresponding to the character;

The character and the first identifier corresponding to the character are input into a pre-generated target polyphone disambiguation model, and the pronunciation of the target polyphone character is determined according to the output of the target polyphone disambiguation model.

Optionally, before the step of dividing the text to be processed into several characters, wherein the several characters include target polyphonic characters and non-target polyphonic characters, it also includes:

Pre-generate the target polyphonic word disambiguation model;

Get pending text.

Optionally, the pre-generated target polyphonic word disambiguation model includes:

Acquire a training sample, wherein the training sample includes several sample texts and annotation information of target polyphonic characters in the sample texts, where the annotation information is used to indicate the pronunciation of the target polyphonic characters in the sample text and the The second identification corresponding to the pronunciation of the target polyphonic character in the sample text;

Taking the several sample texts as input, using the annotation information of the target polyphonic characters in the several sample texts as the output target, training the preset initial model, and determining the trained model as the target polyphonic disambiguation model .

Optionally, determining the pronunciation of the target polyphonic character according to the output of the target polyphonic disambiguation model, including:

According to the output of the target polyphone disambiguation model, the label information of the target polyphone character is obtained;

The pronunciation of the target polyphonic character is determined according to the labeling information of the target polyphonic character.

According to a second aspect of the embodiments of the present application, a device for disambiguating a polyphonic word is provided, the device comprising:

A division module, for dividing the text to be processed into several characters, wherein the several characters include target polyphonic characters and non-target polyphonic characters;

a first identification obtaining module, configured to obtain, for each character, the first identification corresponding to the character;

A polyphonic word disambiguation module, for inputting the character and the first identification corresponding to the character into a pre-generated target polyphonic word disambiguation model, and determining the target polyphonic word according to the output of the target polyphonic word disambiguation model pronunciation of characters.

Optionally, the device further includes:

A polyphonic word disambiguation model training module, which is used to generate a target polyphonic word disambiguation model in advance;

The pending text obtaining module is used to obtain the pending text.

Optionally, the polyphonic word disambiguation model training module includes:

A training sample obtaining unit is configured to obtain a training sample, wherein the training sample includes a number of sample texts and label information of target polyphonic characters in the sample text, and the label information is used to indicate that there are more targets in the sample text. The pronunciation of the phonetic character and the second identification corresponding to the pronunciation of the target polyphonic character in the sample text;

The polyphonic word disambiguation model training unit is used for taking the several sample texts as input, and using the annotation information of the target polyphonic characters in the several sample texts as the output target, training the preset initial model, and completing the training The model is identified as the target polyphonic word disambiguation model.

Optionally, the polyphonic word disambiguation module also includes:

a labeling information obtaining unit, configured to obtain labeling information of the target polyphonic character according to the output of the target polyphonic disambiguation model;

The pronunciation obtaining unit of the target polyphonic character is configured to determine the pronunciation of the target polyphonic character according to the label information of the target polyphonic character.

According to a third aspect of the embodiments of the present application, an electronic device is provided, including:

processor;

memory for storing instructions executable by the processor;

Wherein, the processor is configured to execute the instructions to implement the method for disambiguating polyphonic words.

According to a fourth aspect of the embodiments of the present application, a computer-readable storage medium is provided, when instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device can execute the polyphonic word disambiguation Methods.

The technical solutions provided by the embodiments of the present application may include the following beneficial effects:

In the present application, the text to be processed is divided into several characters, wherein the several characters include target polyphonic characters and non-target polyphonic characters; for each character, the first identifier corresponding to the character is obtained; the The character and the first identifier corresponding to the character are input into a pre-generated target polyphone disambiguation model, and the pronunciation of the target polyphone character is determined according to the output of the target polyphone disambiguation model. Through the technical solutions provided by the embodiments of the present application, the target polyphonic word disambiguation model is used to perform polyphonic word disambiguation on the target polyphonic word character in the text to be processed, thereby improving the prediction speed of polyphonic word disambiguation, and further, improving the multi-phonic word disambiguation. The effect of phonetic disambiguation.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not limiting of the present application.

Description of drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description serve to explain the principles of the invention.

1 is a flowchart of a method for disambiguating a polyphonic word according to an exemplary embodiment;

Fig. 2 is a flowchart of another polyphonic word disambiguation method shown according to an exemplary embodiment;

3 is a flowchart of step 201 in the flowchart of another polyphonic word disambiguation method shown in FIG. 2 according to an exemplary embodiment;

4 is a flowchart of step 103 in the flowchart of a polyphonic word disambiguation method shown in FIG. 1 according to an exemplary embodiment;

5 is a block diagram of a device for disambiguating polyphonic characters according to an exemplary embodiment;

Fig. 6 is another apparatus block diagram of polyphonic word disambiguation shown according to an exemplary embodiment;

Fig. 7 is the device block diagram of the polyphonic word disambiguation model training module 601 in the device block diagram of another polyphonic word disambiguation shown according to an exemplary embodiment shown in Fig. 6;

Fig. 8 is the device block diagram of the polyphonic word disambiguation module 503 in the device block diagram of a polyphonic word disambiguation shown according to an exemplary embodiment shown in Fig. 5;

Fig. 9 is a block diagram of an electronic device according to an exemplary embodiment.

Detailed ways

Exemplary embodiments will be described in detail herein, examples of which are illustrated in the accompanying drawings. Where the following description refers to the drawings, the same numerals in different drawings refer to the same or similar elements unless otherwise indicated. The implementations described in the illustrative examples below are not intended to represent all implementations consistent with the present invention. Rather, they are merely examples of apparatus and methods consistent with some aspects of the invention as recited in the appended claims.

It should be noted that, in the embodiment of the present application, the target polyphonic word disambiguation model may preferably be the fully convolutional model Unet. The reason why the fully convolutional model Unet is selected is because the fully convolutional model Unet can replace the existing The recurrent neural network model RNN used in the technology can achieve better polyphonic word disambiguation effect. The fully convolutional model Unet is most often used in the field of image segmentation. Specifically, the fully convolutional model Unet is named after its symmetrical U-shaped structure, with the convolutional layer on the left and the upsampling layer on the right. The feature maps obtained by each convolutional layer will be spliced to the corresponding upsampling layer to ensure that the features obtained by each layer can be used in subsequent calculations. This also enables the fully convolutional model Unet to combine the features of each layer, thereby improving the polyphony disambiguation model's grasp of the overall features, thereby improving the effect of the polyphonic disambiguation model. The present application uses the fully convolutional model Unet as an embodiment of the target polyphonic word disambiguation model. In the actual application process, other fully convolutional models can also be used to apply to the technical solutions of the present application, for example: including but not limited to IDCNN.

Fig. 1 is a flowchart of a method for disambiguating a polyphonic word according to an exemplary embodiment. As shown in Fig. 1 , the method includes the following steps.

Step 101: Divide the text to be processed into several characters, wherein the several characters include target polyphonic characters and non-target polyphonic characters.

It should be noted that, in the embodiment of the present application, from the perspective of the overall task, polyphonic word disambiguation is a sequence-to-sequence labeling task; however, for each character, it is a classification task. . For example: the pronunciation options of "行" are xing2 or hang2, where 2 refers to the tone of the character being pronounced as yangping (the second tone); the pronunciation options of "zhuan" are zhuan2, zhuan3 and zhuan4, where 3 refers to The tone of the character's pronunciation is the upper tone (the third tone), and 4 means the tone of the character's pronunciation is the go-tone (the fourth tone).

Therefore, firstly, the text to be processed can be divided into characters one by one, specifically, for example: take "login to the People's Bank Credit Information System" as the text to be processed, divide it into characters one by one, and get "login, log in, person, Conduct, sign, trust, style, and standardize".

Further, in this embodiment of the present application, FIG. 2 is a flowchart of another method for disambiguating polyphonic characters according to an exemplary embodiment. As shown in FIG. 2 , the following steps may be further included before step 101 .

Step 201, pre-generate a target polyphonic word disambiguation model.

Further, in the embodiment of the present application, FIG. 3 is a flowchart of step 201 in the flowchart of another polyphonic word disambiguation method shown in FIG. 2 according to an exemplary embodiment, as shown in FIG. 3 , Step 201 may also include the following steps.

Step 301: Obtain training samples, wherein the training samples include several sample texts and label information of target polyphonic characters in the sample texts, where the labeling information is used to indicate the pronunciation of the target polyphonic characters in the sample texts and the second identifier corresponding to the pronunciation of the target polyphonic character in the sample text.

Step 302, using the several sample texts as input, using the label information of the target polyphonic characters in the several sample texts as the output target, training the preset initial model, and determining the model after training as the target polyphonic character Disambiguation model.

It should be noted that, in the embodiment of the present application, the preset initial model can be trained according to the training sample, and the model after training is the target polyphonic word disambiguation model, wherein the target polyphonic word disambiguation model is a full convolution Model Unet structure. Specifically, several sample texts contained in the acquired training samples are used as the input of the initial model, and the label information of the target polyphonic characters in the several sample texts is used as the output target of the initial model, and the preset initial model is trained, and the The trained model is determined as the target polyphonic word disambiguation model, wherein the target polyphonic word disambiguation model is a fully convolutional model Unet. For example: for the pronunciation of the polyphonic character "行" is hang2, get 10,000 sentences containing the polyphonic character "行", among which, in the ten thousand sentences containing the polyphonic character "行", it is necessary to include the word "行" Various pronunciations of characters in natural language, such as: hang2, xing2; in the text with the pronunciation of hang2, adding other characters that are not the polysyllabic character "line" as 0 will be the polysyllabic character "line", and The character whose pronunciation is hang2 obtains its corresponding labeling information according to the preset polyphonic word list, such as: hang2_4, wherein 2 refers to the tone of the pronunciation of the character as yangping (the second tone), and 4 refers to the character Identification in the polyphonic list. Take 10,000 sentences containing the polyphonic character "line" and the text with the pronunciation of hang2 as the input of the initial model, and the label information corresponding to the character, namely hang2_4, as the target of the output, train the initial model, and the training is completed. The model is the target polyphonic word disambiguation model. The application does not specifically limit the form of the marked information and the specific value of the marking.

The present application pre-trains a target polyphonic word disambiguation model, wherein the target polyphonic word disambiguation model is a full convolution model Unet, and then the pre-trained full convolution model Unet can be applied to the method for polyphonic word disambiguation to achieve Improve the efficiency of polyphonic word disambiguation.

Further, in this embodiment of the present application, FIG. 2 is a flowchart of another speech recognition method according to an exemplary embodiment. As shown in FIG. 2 , the following steps may be further included before step 101 .

Step 202, acquiring the text to be processed.

It should be noted that, in this embodiment of the present application, the text to be processed may be any text, which may include at least one target polyphonic character. The text to be processed may be acquired in any manner, and the application does not specifically limit the manner of acquiring the text to be processed.

Step 102: For each character, obtain a first identifier corresponding to the character.

It should be noted that, in this embodiment of the present application, for each character divided by the text to be processed, a first identifier is added to each character, specifically, for example: each divided character "login, land, Person, line, sign, letter, body, system" add the first logo, you can get "Deng_1, Lu_2, person_3, line_4, sign_5, letter_6, system_7, system _8", this application does not specifically limit the specific value of the first identification, and does not specifically limit the form of adding the first identification.

Step 103: Input the character and the first identifier corresponding to the character into a pre-generated target polyphone disambiguation model, and determine the pronunciation of the target polyphone character according to the output of the target polyphone disambiguation model.

Further, in this embodiment of the present application, FIG. 4 is a flowchart of step 103 in the flowchart of a speech recognition method shown in FIG. 1 according to an exemplary embodiment. As shown in FIG. 4 , step 103 further The following steps can be included.

Step 401: Obtain label information of the target polyphone character according to the output of the target polyphone disambiguation model.

Step 402: Determine the pronunciation of the target polyphonic character according to the labeling information of the target polyphonic character.

It should be noted that, in the embodiment of the present application, according to the target polyphonic disambiguation model, that is, the output of the full convolution model Unet, the labeling information of the target polyphonic character is obtained, wherein the labeling information includes the target polyphonic character. The pronunciation corresponds to the second identification of the target polyphonic character, and further, the pronunciation of the target polyphonic character can be determined. For example, if the label information of the polyphonic character "row" is obtained as xing2_8, the corresponding identification of the character in the polyphonic list can be obtained as 8, and its pronunciation is xing yangping (the second tone).

In the present application, the text to be processed is divided into several characters, wherein the several characters include target polyphonic characters and non-target polyphonic characters; for each character, the first identifier corresponding to the character is obtained; the The character and the first identifier corresponding to the character are input into a pre-generated target polyphone disambiguation model, and the pronunciation of the target polyphone character is determined according to the output of the target polyphone disambiguation model. Through the technical solutions provided by the embodiments of the present application, the target polyphone disambiguation model, that is, the full convolution model Unet, is used to disambiguate the target polyphone characters in the text to be processed. Since the full convolution model Unet is easier to parallelize Therefore, there is no sequential relationship between the outputs. By using the fully convolutional model Unet, the prediction speed of polyphonic word disambiguation can not only be improved, but also the context information can be well grasped. Further, the effect of polyphonic word disambiguation is improved. The present application generates a target polyphonic word disambiguation model in advance, wherein the target polyphonic word disambiguation model is a full convolution model Unet, and then the pre-trained full convolution model Unet can be applied to the method for polyphonic word disambiguation, and further Therefore, the processing efficiency and prediction accuracy of polyphonic word disambiguation are improved.

FIG. 5 is a block diagram of an apparatus for disambiguating a polyphonic word according to an exemplary embodiment. Referring to FIG. 5 , the apparatus includes a division module 501 , a first identification obtaining module 502 , and a polyphonic word disambiguation module 503 .

The dividing module 501 is configured to divide the text to be processed into several characters, wherein the several characters include target polyphonic characters and non-target polyphonic characters.

The first identification obtaining module 502 is configured to obtain, for each character, the first identification corresponding to the character.

The polyphonic word disambiguation module 503 is configured to input the character and the first identification corresponding to the character into the pre-generated target polyphonic word disambiguation model, and determine the target polyphonic word disambiguation model according to the output of the target polyphonic word disambiguation model. Pronunciation of phonetic characters.

Optionally, FIG. 6 is a block diagram of another apparatus for disambiguating polyphonic characters according to an exemplary embodiment. Referring to FIG. 6 , the apparatus includes a polyphonic word disambiguation model training module 601 and a to-be-processed text acquisition module 602 .

The polyphonic word disambiguation model training module 601 is used to generate a target polyphonic word disambiguation model in advance.

The to-be-processed text acquisition module 602 is configured to acquire the to-be-processed text.

Optionally, FIG. 7 is an apparatus block diagram of a polyphonic word disambiguation model training module 601 in another apparatus block diagram of polyphonic word disambiguation shown in FIG. 6 according to an exemplary embodiment. Referring to FIG. 7 , the apparatus includes a training sample acquisition unit 701 and a polyphonic word disambiguation model training unit 702 .

A training sample obtaining unit 701, configured to obtain a training sample, wherein the training sample includes several sample texts and several annotation information of target polyphonic characters in the sample text, and the annotation information is used to indicate the target in the sample text. The pronunciation of the polyphonic character and the second identifier corresponding to the pronunciation of the target polyphonic character in the sample text.

The polyphonic word disambiguation model training unit 702 is configured to use the several sample texts as input, and use the label information of the target polyphonic word characters in the several sample texts as the output target, train a preset initial model, and train the training The completed model is determined as the target polyphonic word disambiguation model.

Optionally, FIG. 8 is an apparatus block diagram of the polyphonic word disambiguation module 503 in the apparatus block diagram of a polyphonic word disambiguation shown in FIG. 5 according to an exemplary embodiment. Referring to FIG. 8 , the apparatus includes an annotation information acquisition unit 801 and a pronunciation acquisition unit 802 of a target polyphonic character.

The labeling information obtaining unit 801 is configured to obtain labeling information of the target polyphonic character according to the output of the target polyphonic disambiguation model.

The pronunciation obtaining unit 802 of the target polyphonic character is configured to determine the pronunciation of the target polyphonic character according to the label information of the target polyphonic character.

In the present application, the text to be processed is divided into several characters, wherein the several characters include target polyphonic characters and non-target polyphonic characters; for each character, the first identifier corresponding to the character is obtained; the The character and the first identifier corresponding to the character are input into a pre-generated target polyphone disambiguation model, and the pronunciation of the target polyphone character is determined according to the output of the target polyphone disambiguation model. Through the technical solutions provided by the embodiments of the present application, the target polyphone disambiguation model, that is, the full convolution model Unet, is used to disambiguate the target polyphone characters in the text to be processed. Since the full convolution model Unet is easier to parallelize Therefore, there is no sequential relationship between the outputs. By using the fully convolutional model Unet, the prediction speed of polyphonic word disambiguation can not only be improved, but also the context information can be well grasped. Further, the effect of polyphonic word disambiguation is improved. The present application generates a target polyphonic word disambiguation model in advance, wherein the target polyphonic word disambiguation model is a full convolution model Unet, and then the pre-trained full convolution model Unet can be applied to the method for polyphonic word disambiguation, and further Therefore, the processing efficiency and prediction accuracy of polyphonic word disambiguation are improved.

Regarding the apparatus in the above-mentioned embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be described in detail here.

FIG. 9 is a block diagram of an electronic device 900 according to an exemplary embodiment. For example, electronic device 900 may be a mobile phone, computer, digital broadcast terminal, messaging device, game console, tablet device, medical device, fitness device, personal digital assistant, and the like.

9, electronic device 900 may include one or more of the following components: processing component 902, memory 904, power supply component 906, multimedia component 908, audio component 910, input/output interface 912, sensor component 914, and communication component 916.

The processing component 902 generally controls the overall operation of the apparatus 900, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 902 may include one or more processors 920 to execute instructions to perform all or some of the steps of the methods described above. Additionally, processing component 902 may include one or more modules to facilitate interaction between processing component 902 and other components. For example, processing component 902 may include a multimedia module to facilitate interaction between multimedia component 908 and processing component 902.

Memory 904 is configured to store various types of data to support operation at device 900 . Examples of such data include instructions for any application or method operating on device 900, contact data, phonebook data, messages, pictures, videos, and the like. Memory 904 may be implemented by any type of volatile or non-volatile storage device or combination thereof, such as static random access memory (SRAM), electrically erasable programmable read only memory (EEPROM), erasable Programmable Read Only Memory (EPROM), Programmable Read Only Memory (PROM), Read Only Memory (ROM), Magnetic Memory, Flash Memory, Magnetic or Optical Disk.

Power supply assembly 906 provides power to various components of electronic device 900 . Power supply components 906 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to electronic device 900 .

Multimedia component 908 includes a screen that provides an output interface between the electronic device 900 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from a user. The touch panel includes one or more touch sensors to sense touch, swipe, and gestures on the touch panel. The touch sensor may not only sense the boundaries of a touch or swipe action, but also detect the duration and pressure associated with the touch or swipe action. In some embodiments, the multimedia component 908 includes a front-facing camera and/or a rear-facing camera. When the electronic device 900 is in an operation mode, such as a shooting mode or a video mode, the front camera and/or the rear camera may receive external multimedia data. Each of the front and rear cameras can be a fixed optical lens system or have focal length and optical zoom capability.

Audio component 910 is configured to output and/or input audio signals. For example, audio component 910 includes a microphone (MIC) that is configured to receive external audio signals when electronic device 900 is in operating modes, such as calling mode, recording mode, and voice recognition mode. The received audio signal may be further stored in memory 904 or transmitted via communication component 916 . In some embodiments, audio component 910 also includes a speaker for outputting audio signals.

The input/output interface 912 provides an interface between the processing component 902 and a peripheral interface module, which may be a keyboard, a click wheel, a button, or the like. These buttons may include, but are not limited to: home button, volume buttons, start button, and lock button.

Sensor assembly 914 includes one or more sensors for providing status assessments of various aspects of electronic device 900 . For example, the sensor assembly 914 can detect the open/closed state of the electronic device 900, the relative positioning of the components, such as the display and the keypad of the electronic device 900, the sensor assembly 914 can also detect the electronic device 900 or one of the electronic devices 900 Changes in the position of components, presence or absence of user contact with the electronic device 900 , orientation or acceleration/deceleration of the electronic device 900 and changes in the temperature of the electronic device 900 . Sensor assembly 914 may include a proximity sensor configured to detect the presence of nearby objects in the absence of any physical contact. Sensor assembly 914 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 914 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

Communication component 916 is configured to facilitate wired or wireless communication between electronic device 900 and other devices. Electronic device 900 may access wireless networks based on communication standards, such as WiFi, carrier networks (eg, 2G, 3G, 4G, or 5G), or a combination thereof. In one exemplary embodiment, the communication component 916 receives broadcast signals or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 1116 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

In an exemplary embodiment, electronic device 900 may be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable A programmed gate array (FPGA), controller, microcontroller, microprocessor or other electronic component implementation is used to perform the above method.

In an exemplary embodiment, there is also provided a non-transitory computer-readable storage medium including instructions, such as memory 904 including instructions, executable by the processor 920 of the electronic device 900 to perform the above method. For example, the non-transitory computer-readable storage medium may be ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, optical data storage device, and the like.

Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. The present invention is intended to cover any modifications, uses or adaptations of the present invention which follow the general principles of the invention and include common knowledge or common technical means in the technical field not disclosed in this application . The specification and examples are to be regarded as exemplary only, with the true scope and spirit of the invention being indicated by the following claims.

It should be understood that the present invention is not limited to the precise structures described above and illustrated in the accompanying drawings, and that various modifications and changes may be made without departing from its scope. The scope of the present invention is limited only by the appended claims.

Claims (10)

1. A method of polyphonic disambiguation, the method comprising:

dividing a text to be processed into a plurality of characters, wherein the plurality of characters comprise target polyphone characters and non-target polyphone characters;

aiming at each character, acquiring a first identifier corresponding to the character;

inputting the characters and the first identifications corresponding to the characters into a pre-generated target polyphonic character disambiguation model, and determining the pronunciation of the target polyphonic character according to the output of the target polyphonic character disambiguation model.

2. The method of claim 1, wherein before the step of dividing the text to be processed into a number of characters, wherein the number of characters includes a target polyphonic character and a non-target polyphonic character, further comprising:

generating a target polyphone disambiguation model in advance;

and acquiring a text to be processed.

3. The method of claim 2, wherein the pre-generating a target polyphonic disambiguation model comprises:

acquiring a training sample, wherein the training sample comprises a plurality of sample texts and marking information of target polyphonic characters in the plurality of sample texts, and the marking information is used for indicating pronunciation of the target polyphonic characters in the sample texts and second identification corresponding to the pronunciation of the target polyphonic characters in the sample texts;

and taking the sample texts as input, taking the marking information of the target polyphone characters in the sample texts as an output target, training a preset initial model, and determining the trained model as a target polyphone disambiguation model.

4. The method of claim 1, wherein determining the pronunciation of the target polyphonic character from the output of the target polyphonic disambiguation model comprises:

obtaining the marking information of the target polyphone character according to the output of the target polyphone disambiguation model;

and determining the pronunciation of the target polyphonic character according to the labeling information of the target polyphonic character.

5. An apparatus for polyphonic disambiguation, the apparatus comprising:

the device comprises a dividing module, a processing module and a processing module, wherein the dividing module is used for dividing a text to be processed into a plurality of characters, and the plurality of characters comprise target polyphonic characters and non-target polyphonic characters;

the first identification acquisition module is used for acquiring a first identification corresponding to each character;

and the polyphone disambiguation module is used for inputting the characters and the first identifications corresponding to the characters into a pre-generated target polyphone disambiguation model and determining the pronunciation of the target polyphone characters according to the output of the target polyphone disambiguation model.

6. The apparatus of claim 5, further comprising:

the polyphone disambiguation model training module is used for generating a target polyphone disambiguation model in advance;

and the to-be-processed text acquisition module is used for acquiring the to-be-processed text.

7. The apparatus of claim 6, wherein the polyphonic disambiguation model training module comprises:

the training sample acquisition unit is used for acquiring a training sample, wherein the training sample comprises a plurality of sample texts and marking information of target polyphonic characters in the plurality of sample texts, and the marking information is used for indicating pronunciation of the target polyphonic characters in the sample texts and second identification corresponding to the pronunciation of the target polyphonic characters in the sample texts;

and the polyphone disambiguation model training unit is used for taking the plurality of sample texts as input, taking the marking information of the target polyphone characters in the plurality of sample texts as an output target, training a preset initial model, and determining the trained model as the target polyphone disambiguation model.

8. The apparatus of claim 5, wherein the polyphonic disambiguation module further comprises:

the label information acquisition unit is used for acquiring label information of the target polyphonic characters according to the output of the target polyphonic disambiguation model;

and the pronunciation acquisition unit of the target polyphonic character is used for determining the pronunciation of the target polyphonic character according to the labeling information of the target polyphonic character.

9. An electronic device, comprising:

a processor;

a memory for storing processor-executable instructions;

wherein the processor is configured to execute the instructions to implement a method of polyphonic disambiguation as claimed in any of claims 1 to 4.

10. A computer readable storage medium having instructions that, when executed by a processor of a mobile terminal, enable the mobile terminal to perform a method of polyphonic disambiguation as claimed in any of claims 1 to 4.

Images