End-to-end acoustic modeling using convolutional neural networks for HMM-based automatic speech recognition.
Dimitri PalazMathew Magimai-DossRonan CollobertPublished in: Speech Commun. (2019)
Keyphrases
- end to end
- automatic speech recognition
- hidden markov models
- convolutional neural networks
- acoustic features
- speech recognition
- speech sounds
- speech recognizers
- speech signal
- speech segments
- speech retrieval
- broadcast news
- conversational speech
- word error rate
- acoustic models
- congestion control
- speech corpus
- spoken words
- recognition errors
- scalable video
- admission control
- noisy environments
- internet protocol
- word recognition
- video sequences
- transport layer
- application layer
- handwriting recognition
- neural network