Speech enhancement using beamforming and non negative matrix factorization for robust speech recognition in the CHiME-3 challenge.
Thanh T. VuBenjamin BigotEngsiong ChngPublished in: ASRU (2015)
Keyphrases
- speech enhancement
- noisy environments
- speech recognition
- negative matrix factorization
- speech signal
- noisy speech
- speaker identification
- language model
- hidden markov models
- speech synthesis
- noise reduction
- matrix factorization
- automatic speech recognition
- pattern recognition
- document clustering
- background noise
- principal component analysis
- sparse representation
- additive noise
- linear prediction
- feature selection
- natural language
- image segmentation