Likelihood-Maximizing-Based Multiband Spectral Subtraction for Robust Speech Recognition.
Bagher BabaAliHossein SametiMehran SafayaniPublished in: EURASIP J. Adv. Signal Process. (2009)
Keyphrases
- speech recognition
- spectral subtraction
- multi band
- noisy environments
- speech signal
- hidden markov models
- multispectral
- frequency band
- automatic speech recognition
- language model
- speech synthesis
- noisy speech
- pattern recognition
- multiscale
- speech enhancement
- maximum likelihood
- denoising
- noise removal
- image data
- probabilistic model
- speaker identification
- speaker verification
- image processing
- feature selection