Nonnegative matrix decomposition method for speech signal characteristic waveform
A technology of non-negative matrix decomposition and characteristic waveform, which is applied in the field of speech signal processing and can solve problems such as low complexity and high precision
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0085] In the specific implementation manner of the present invention, how to use the standard non-negative matrix decomposition method to train the base matrix and how to obtain the coding matrix are respectively discussed below.
[0086] A. Training base matrix.
[0087] a. At first, the voice feature waveform is divided into 9 categories according to the size of the pitch period (pitch) of the frame voice signal, and the classification basis is as shown in Table 1:
[0088] Class 1
20≤pitch<30
Class 2
30≤pitch<40
Class 3
40≤pitch<50
Class 4
50≤pitch<60
Class 5
60≤pitch<70
Class 6
70≤pitch<80
Class 7
80≤pitch<90
Class 8
90≤pitch<100
Class 9
100≤pitch≤120
[0089] Table 1 Classification of characteristic waveforms
[0090] B, then all select about 10000 frames of experimental samples for each type of characteristic waveform, form matrix V, according to the iterative ...
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