Punctuation predicting, labeling and voice processing method, device, equipment and program product
A prediction processing and punctuation technology, applied in the computer field, can solve the problems of low punctuation prediction accuracy, inability to explicitly model punctuation, slow decoding speed, etc., and achieve the effect of improving the punctuation prediction rate
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0054] like image 3 As shown, it is a schematic flowchart of a punctuation prediction processing method according to an embodiment of the present invention. The method can be applied to a tool for punctuation processing of text, and can also be applied to a speech recognition model. Voice punctuation. Specifically, it can be applied to the server side or the terminal side, and the method can include:
[0055] S101: Acquire text feature data after semantic encoding of the text to be processed. The text to be processed may be any text that has not been punctuated, or of course, text that has been partially punctuated, and the length of the text to be processed is not limited. The punctuation prediction processing method in the embodiment of the present invention can be implemented by using the structure of an encoder and a decoder, and specifically, an RNN structure can be used to implement the encoder and the decoder respectively, or an attention mechanism-based Transfo...
Embodiment 2
[0069] like Figure 4 As shown, it is a schematic structural diagram of a punctuation prediction processing device according to an embodiment of the present invention. The device can be applied to a tool for punctuation processing of text, and can also be applied to a speech recognition model for identifying Voice punctuation. Specifically, it can be applied to the server side or the terminal side, and the device can include:
[0070] The text feature data acquisition module 11 is used for acquiring the text feature data after semantic encoding of the text to be processed. The text to be processed may be any text that has not been punctuated, or of course, text that has been partially punctuated, and the length of the text to be processed is not limited. The punctuation prediction processing apparatus according to the embodiment of the present invention can be implemented by using the structure of an encoder and a decoder. Specifically, an RNN structure can be used to i...
Embodiment 3
[0083] Embodiments of the present invention also provide a speech recognition method, which can be applied to speech recognition scenarios, such as Figure 5 and Image 6 As shown, it is a schematic diagram of an application scenario of the speech recognition method according to the embodiment of the present invention. Figure 5 and Image 6 The schematic diagram to illustrate the processing process of the above-mentioned speech recognition method, the processing process includes:
[0084] S201: Recognize input speech, and generate speech recognition text. like Figure 5 and Image 6 As shown in the figure, the input speech may be collected from some APPs with speech recognition function, and then the collected input speech is recognized by the speech recognition model to generate speech recognition text. Among them, speech recognition can be done on the client side, such as Figure 5 As shown in the figure, the speech recognition model is deployed on the clie...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com