Self-regularised Minimum Latency Training for Streaming Transformer-based Speech Recognition.
Mohan LiRama Sanand DoddipatlaCatalin ZorilaPublished in: INTERSPEECH (2022)
Keyphrases
- speech recognition
- wall street journal corpus
- isolated word
- hidden markov models
- acoustic models
- language model
- speech processing
- speech signal
- speech synthesis
- automatic speech recognition
- pattern recognition
- speaker identification
- speech recognizer
- speech recognition technology
- speech recognizers
- noisy environments
- keyword spotting
- speaker independent
- speech retrieval
- cepstral coefficients
- speech recognition errors
- training process