wav2vec: Unsupervised Pre-Training for Speech Recognition.
Steffen SchneiderAlexei BaevskiRonan CollobertMichael AuliPublished in: INTERSPEECH (2019)
Keyphrases
- speech recognition
- wall street journal corpus
- isolated word
- supervised learning
- hidden markov models
- automatic speech recognition
- speech processing
- acoustic models
- pattern recognition
- speech signal
- language model
- speech recognizer
- speech recognition technology
- speech synthesis
- speaker identification
- training process
- speech recognition systems
- cepstral coefficients
- unsupervised learning
- speech understanding
- speaker diarization
- semi supervised
- speech recognition errors
- training set
- keyword spotting
- speaker independent
- neural network