Wind Noise Reduction using Non-Negative Sparse Coding
- 1 August 2007
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE) in 2007 IEEE Workshop on Machine Learning for Signal Processing
- No. 15512541,p. 431-436
Abstract
We introduce a new speaker independent method for reducing wind noise in single-channel recordings of noisy speech. The method is based on non-negative sparse coding and relies on a wind noise dictionary which is estimated from an isolated noise recording. We estimate the parameters of the model and discuss their sensitivity. We then compare the algorithm with the classical spectral subtraction method and the Qualcomm-ICSI-OGI noise reduction method. We optimize the sound quality in terms of signal-to-noise ratio and provide results on a noisy speech recognition task.Keywords
This publication has 10 references indexed in Scilit:
- Projected Gradient Methods for Nonnegative Matrix FactorizationNeural Computation, 2007
- Fast Newton-type Methods for the Least Squares Nonnegative Matrix Approximation Problem2007
- Model-Based Monaural Source Separation Using a Vector-Quantized Phase-Vocoder Representation2006
- Multilayer nonnegative matrix factorisationElectronics Letters, 2006
- Enhancement of speech corrupted by acoustic noise2005
- Sparse coding and NMF2005
- Factorial models and refiltering for speech separation and denoising2003
- Non-negative sparse coding2003
- Learning the parts of objects by non-negative matrix factorizationNature, 1999
- Suppression of acoustic noise in speech using spectral subtractionIEEE Transactions on Acoustics, Speech, and Signal Processing, 1979