Single-sided speech quality measurement
US9786300B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | Aug 1, 2011 |
| Grant date | Oct 10, 2017 |
| Priority date | — |
| Expiry date | Oct 19, 2032 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG10L25/69
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
A non-intrusive speech quality estimation technique is based on statistical or probability models such as Gaussian Mixture Models (“GMMs”). Perceptual features are extracted from the received speech signal and assessed by an artificial reference model formed using statistical models. The models characterize the statistical behavior of speech features. Consistency measures between the input speech features and the models are calculated to form indicators of speech quality. The consistency values are mapped to a speech quality score using a mapping optimized using machine learning algorithms, such as Multivariate Adaptive Regression Splines (“MARS”). The technique provides competitive or better quality estimates relative to known techniques while having lower computational complexity.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.