System and Method for Dynamic Noise Adaptation for Robust Automatic Speech Recognition
AssigneeNuance Communications, Inc.
InventorsSteven J. Rennie, Pierre Dognin, Petr Fousek
Life of this patent
Estimating the term…
Abstract
A speech processing method and arrangement are described. A dynamic noise adaptation (DNA) model characterizes a speech input reflecting effects of background noise. A null noise DNA model characterizes the speech input based on reflecting a null noise mismatch condition. A DNA interaction model performs Bayesian model selection and re-weighting of the DNA model and the null noise DNA model to realize a modified DNA model characterizing the speech input for automatic speech recognition and compensating for noise to a varying degree depending on relative probabilities of the DNA model and the null noise DNA model.
Claims21
A computer-implemented method for automatic speech recognition comprising:
- characterizing a speech input in a computer process based on a dynamic noise adaptation (DNA) model reflecting effects of background noise;
- characterizing the speech input in a computer process based on a null noise DNA model reflecting a null noise mismatch condition;
- performing in a computer process Bayesian model selection and re-weighting of the DNA model and the null noise DNA model to realize a modified DNA model characterizing the speech input for automatic speech recognition and compensating for noise to a varying degree depending on relative probabilities of the DNA model and the null noise DNA model.
A speech processing arrangement employing at least one hardware implemented computer processor and comprising:
- a dynamic noise adaptation (DNA) model configured to electronically characterize a speech input reflecting effects of background noise;
- a null noise DNA model configured to electronically characterize the speech input based on reflecting a null noise mismatch condition;
- a DNA interaction model configured to electronically perform Bayesian model selection and re-weighting of the DNA model and the null noise DNA model to realize a modified DNA model characterizing the speech input for automatic speech recognition and compensating for noise to a varying degree depending on relative probabilities of the DNA model and the null noise DNA model.
A computer program product encoded in a non-transitory computer-readable medium for speech processing, the product comprising:
- program code for characterizing a speech input based on a dynamic noise adaptation (DNA) model reflecting effects of background noise;
- program code for characterizing the speech input based on a null noise DNA model reflecting a null noise mismatch condition;
- program code for performing Bayesian model selection and re-weighting of the DNA model and the null noise DNA model to realize a modified DNA model characterizing the speech input for automatic speech recognition and compensating for noise to a varying degree depending on relative probabilities of the DNA model and the null noise DNA model.
Dependencies are inferred from the claim text.