Method for improving rejection capability of speech recognition system
A technology of speech recognition and ability, applied in speech recognition, speech analysis, instruments, etc., can solve the problems of high algorithm complexity, poor universality, and difficulty in successful application, so as to achieve uncomplicated calculation, improve robustness, Improve the effect of denial effect
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0029] Such as figure 2 As shown, the present invention may improve the ability to reject invalid input such as out-of-set words, background voices, and other noises. The specific process is as follows:
[0030] (1) Collect a variety of noise data; then classify according to the type of noise, including background noise, background music, door closing and coughing; then train Gaussian mixture models (GMM) for different types of noise; finally combine All kinds of GMM models are the overall absorption model; the Gaussian mixture model GMM (Gaussian mixture model) is an extension of a single Gaussian density function, which can smoothly approximate the density distribution of any shape, which is why the GMM model is often used in the field of speech recognition. one;
[0031] (2) A statistical language model is trained through a variety of relatively random texts, and then a recognition network is constructed through weighted finite state machine (WFST) technology, which is ca...
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