Modeling the dynamics of speech and noise for speech feature enhancement in ASR.
Stefan WindmannReinhold Haeb-UmbachPublished in: ICASSP (2008)
Keyphrases
- speech recognition
- automatic speech recognition
- noisy environments
- speech signal
- speech enhancement
- spontaneous speech
- broadcast news
- speech synthesis
- additive noise
- background noise
- recognition engine
- hidden markov models
- pattern recognition
- speaker identification
- signal to noise ratio
- spoken words
- speech corpus
- spoken language
- endpoint detection
- speech recognizers
- speech recognizer
- conversational speech
- random noise
- feature vectors
- recognition errors
- word error rate
- cepstral features
- image processing
- language model
- vocal tract
- image enhancement
- speaker recognition
- noise reduction
- language acquisition