Effect of speech priors in single-channel speech-music separation for ASR.
Cemil DemirAli Taylan CemgilMurat SaraçlarPublished in: INTERSPEECH (2012)
Keyphrases
- automatic speech recognition
- speech recognition
- speech enhancement
- single channel
- speech signal
- noisy environments
- multi channel
- spontaneous speech
- text to speech
- audio features
- sound source
- speech corpus
- prior information
- noise reduction
- vocal tract
- hidden markov models
- independent component analysis
- audio visual
- signal to noise ratio
- linear prediction
- non stationary
- motion estimation
- pattern recognition
- feature selection