A kind of speech enhancement method and system
A technology of speech enhancement and speech characteristics, applied in speech analysis, instruments, etc., can solve the problems of poor speech perception quality and low intelligibility, and achieve good speech enhancement effect
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0035] The embodiment of the present invention provides a speech enhancement method, which can be applied to scenarios such as cochlear implants, hearing aids, human-computer interaction systems, and speech communication, such as figure 1 As shown, the method includes the following steps:
[0036] Step S1: Build a speech enhancement network model, the network model includes three sub-neural networks, wherein the first neural network is a common part, and it and the second neural network constitute a prediction time-frequency mask module, and at the same time constitute a prediction module with the third neural network Adaptive weight module.
[0037] In the embodiment of the present invention, the constructed neural network model includes two parallel modules, wherein the predictive adaptive weight module judges the signal-to-noise ratio according to the input characteristics, thereby adjusting the proportion of speech distortion and residual noise through the weight, and pred...
Embodiment 2
[0051] An embodiment of the present invention provides a speech enhancement system, such as Figure 4 shown, including:
[0052] Model construction module 1, is used for constructing speech enhancement network model, and described network model comprises three sub-neural networks, and wherein the first neural network is a common part, and it and the second neural network constitute the time-frequency mask module of prediction, simultaneously and the third The neural network constitutes a predictive adaptive weight module; this module executes the method described in step S1 in Embodiment 1, which will not be repeated here.
[0053] Model training module 2, is used for inputting the speech characteristic of band noise speech signal in described network model, and the first neural network generates an intermediate latent variable according to the speech characteristic of input, and described intermediate latent variable simultaneously serves as the second neural network and the ...
Embodiment 3
[0057] An embodiment of the present invention provides a computer device, such as Figure 5 As shown, the device may include a processor 51 and a memory 52, wherein the processor 51 and the memory 52 may be connected via a bus or in other ways, Figure 5 Take connection via bus as an example.
[0058] As a non-transitory computer-readable storage medium, the memory 52 can be used to store non-transitory software programs, non-transitory computer-executable programs and modules, such as corresponding program instructions / modules in the embodiments of the present invention. The processor 51 executes various functional applications and data processing of the processor by running the non-transitory software programs, instructions and modules stored in the memory 52, that is, implements the speech enhancement method in the first method embodiment above.
[0059] The memory 52 may include a program storage area and a data storage area, wherein the program storage area may store an ...
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