Noise adaptation of HMM speech recognition systems using tied-mixtures in the spectral domain.
Adoram ErellDavid BurshteinPublished in: IEEE Trans. Speech Audio Process. (1997)
Keyphrases
- speech recognition
- speech recognition systems
- spectral domain
- hidden markov models
- noisy environments
- frequency domain
- speech recognizer
- spatial domain
- speech signal
- automatic speech recognition
- pattern recognition
- spectral analysis
- optical coherence tomography
- language model
- noise reduction
- blind source separation
- mixture model
- information retrieval
- signal to noise ratio
- probabilistic model
- multiscale
- feature selection