Joint training of DNNs by incorporating an explicit dereverberation structure for distant speech recognition.
Tian GaoJun DuYong XuCong LiuLi-Rong DaiChin-Hui LeePublished in: EURASIP J. Adv. Signal Process. (2016)
Keyphrases
- speech recognition
- wall street journal corpus
- hidden markov models
- speech synthesis
- isolated word
- language model
- automatic speech recognition
- speech recognizer
- speech signal
- speech recognition technology
- speaker identification
- speech processing
- pattern recognition
- speech understanding
- speech recognition errors
- noisy environments
- audio visual speech recognition
- speech recognition systems
- students with learning disabilities
- speaker adaptation
- speaker diarization
- keyword spotting
- speech retrieval
- speech recognizers
- maximum likelihood
- feature selection