阅读说明：本技术 语音识别方法、装置、电子设备和计算机可读存储介质 (Speech recognition method, speech recognition device, electronic equipment and computer-readable storage medium ) 是由 谭志鹏 谭北平 于 2019-12-06 设计创作，主要内容包括：本申请提供一种语音识别方法、装置、电子设备和计算机可读存储介质，涉及语音识别技术领域。该语音识别方法应用于电子设备，电子设备维护有标准语音库和类比预测模型，该语音识别方法包括：根据用户输入的待识别语音信息获取第一特征信息；第一特征信息用于确定电子设备的反馈信息；判断第一特征信息与标准语音库是否匹配；若匹配，则执行与第一特征信息对应的第一目标操作；若不匹配，则根据第一特征信息和类比预测模型，执行第二目标操作。在每次用户语音输入信息后，将语音信息中的特征信息与标准语音库进行比对，比对不成功时，使用类比预测模型补充语音信息中的特征信息，并执行相应的操作，提高用户体验。(The application provides a voice recognition method, a voice recognition device, electronic equipment and a computer readable storage medium, and relates to the technical field of voice recognition. The speech recognition method is applied to electronic equipment, the electronic equipment maintains a standard speech base and an analog prediction model, and the speech recognition method comprises the following steps: acquiring first characteristic information according to voice information to be recognized input by a user; the first characteristic information is used for determining feedback information of the electronic equipment; judging whether the first characteristic information is matched with a standard voice library or not; if so, executing a first target operation corresponding to the first characteristic information; if not, executing a second target operation according to the first characteristic information and the analogy prediction model. After the user inputs information by voice each time, the characteristic information in the voice information is compared with a standard voice library, and when the comparison is unsuccessful, the characteristic information in the voice information is supplemented by using an analog prediction model, and corresponding operation is executed, so that the user experience is improved.)

1. A speech recognition method applied to an electronic device, the electronic device maintaining a standard speech library and an analog prediction model, the analog prediction model being used for determining operation information of the electronic device according to speech characteristics, the method comprising:

acquiring first characteristic information according to voice information to be recognized input by a user; the first characteristic information is used for determining feedback information of the electronic equipment;

judging whether the first characteristic information is matched with the standard voice library or not;

if so, executing a first target operation corresponding to the first characteristic information;

and if not, executing a second target operation according to the first characteristic information and the analog prediction model.

2. The method according to claim 1, wherein the electronic device further maintains a personalized speech library, the personalized speech library comprising a plurality of specific feature information, and before performing a second target operation according to the first feature information and the analogy prediction model, the method further comprises:

and acquiring the analogy prediction model according to the specific characteristic information.

3. The method of claim 2, further comprising:

and storing the first characteristic information to the personalized voice library.

4. The method of claim 2 or 3, wherein the electronic device further maintains an analogy information base, and the obtaining the analogy prediction model according to the specific feature information comprises:

acquiring the relevancy information of the plurality of specific characteristic information;

storing a plurality of training feature information into the analogy information base; the training characteristic information is the specific characteristic information of which the relevancy information meets a preset condition in the plurality of specific characteristic information;

and obtaining the analogy prediction model according to the analogy information base.

5. A speech recognition apparatus, comprising: the device comprises an acquisition module, a judgment module and a control module;

the voice recognition device maintains a standard voice library and an analog prediction model, and the analog prediction model is used for determining the operation information of the voice recognition device according to voice characteristics;

the acquisition module is used for acquiring first characteristic information according to the voice information to be recognized input by a user; the first characteristic information is used for determining feedback information of the voice recognition device;

the judging module is used for judging whether the first characteristic information is matched with the standard voice library or not;

the control module is used for executing a first target operation corresponding to the first characteristic information when the first characteristic information is matched with the standard voice library;

the control module is further configured to execute a second target operation according to the first feature information and the analog prediction model when the first feature information does not match the standard voice library.

6. The apparatus according to claim 5, wherein the speech recognition apparatus further maintains a personalized speech library, the personalized speech library comprising a plurality of specific feature information;

the obtaining module is further configured to obtain the analogy prediction model according to the specific feature information.

7. The apparatus of claim 6, further comprising a storage module;

the storage module is used for storing the first characteristic information to the personalized voice library.

8. The apparatus of claim 6 or 7, wherein the speech recognition apparatus further maintains a library of analogy information;

the obtaining module is further configured to obtain relevancy information of the plurality of specific feature information;

the control module is also used for storing a plurality of training characteristic information to the analogy information base; the training characteristic information is the specific characteristic information of which the relevancy information meets a preset condition in the plurality of specific characteristic information;

the acquisition module is further configured to acquire the analogy prediction model according to the analogy information base.

9. An electronic device comprising a processor and a memory, the memory storing machine executable instructions executable by the processor to implement the speech recognition method of any one of claims 1-4.

10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the speech recognition method according to any one of claims 1 to 4.