Voice recognition method and device and storage medium
A speech recognition and speech recognition model technology, applied in speech recognition, speech analysis, biological neural network models, etc., can solve problems such as lack of data support, lack of accuracy and reliability of speech recognition, and wake-up
Pending Publication Date: 2020-11-20
BEIJING XIAOMI PINECONE ELECTRONICS CO LTD
View PDF9 Cites 2 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
The neural network model is generally composed of multiple sub-modules. In related technologies, each sub-module in the neural network model is manually selected. Since the manual selection method does not have accurate data support, the performance of the generated network model is not good.
Moreover, the neural network model in the related art generally reuses the model in the visual field, which is not effective for speech recognition.
In this way, the accuracy and reliability of speech recognition in the related art are insufficient. For example, when the user wants to wake up the terminal device, the device may not be woken up in time due to the inaccurate recognition of the speech spoken by the user, so the device cannot be woken up quickly. Good to meet the needs of users
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0038] Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numerals in different drawings refer to the same or similar elements unless otherwise indicated. The implementations described in the following exemplary examples do not represent all implementations consistent with the present disclosure. Rather, they are merely examples of apparatuses and methods consistent with aspects of the present disclosure as recited in the appended claims.
[0039] figure 1 is a flow chart of a speech recognition method shown according to an exemplary embodiment, such as figure 1 As shown, the method may include S101 and S102.
[0040] In S101, when voice information is received, the voice information is input into the generated voice recognition model.
[0041] In S102, a recognition result is output through the speech recognition mode...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention relates to a voice recognition method and device and a storage medium. The method comprises the steps that when voice information is received, the voice information is input into a generated voice recognition model; and a recognition result is output through a voice recognition model. The method for generating the voice recognition model comprises the steps that a super network is trained, wherein the super network comprises multiple network layers, each network layer comprises M substructures, at least one substructure in the M substructures comprises a time sequence convolutionnetwork module, and M is a positive integer larger than or equal to 2; a target substructure corresponding to each network layer is determined from the M substructures of the network layer accordingto a training result; and the voice recognition model is generated according to the target substructure corresponding to each network layer. Through the technical scheme, the performance of the voicerecognition model is improved, the accuracy of voice recognition is ensured, and the recognition speed and the response speed of the voice information are improved.
Description
technical field [0001] The present disclosure relates to the field of voice recognition, and in particular, to a voice recognition method, device and storage medium. Background technique [0002] Speech recognition can be simply explained as the recognition of speech or sound signals, and it is widely used in various fields. For example, when starting a terminal device, the user can wake up the terminal device by speaking a short voice without turning on a switch or fingerprint recognition. This method of starting the device is convenient and fast. When the voice wakes up, the device is activated from the dormant state to the running state by detecting voice keywords. The response speed and accuracy of the voice wakeup directly affect the user's experience in using the device. [0003] Currently, speech recognition is usually performed through a neural network model, such as an end-to-end neural network recognition model. A neural network model is generally composed of mul...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G10L15/22G06N3/04G10L15/06G10L15/16
CPCG10L15/22G10L15/063G10L15/16G06N3/045
Inventor 张勃初祥祥李庆源
Owner BEIJING XIAOMI PINECONE ELECTRONICS CO LTD