Self-regularised Minimum Latency Training for Streaming Transformer-based Speech Recognition.
Mohan LiRama DoddipatlaCatalin ZorilaPublished in: CoRR (2023)
Keyphrases
- speech recognition
- wall street journal corpus
- isolated word
- hidden markov models
- language model
- automatic speech recognition
- speech synthesis
- acoustic models
- speech processing
- speech recognizer
- speech signal
- pattern recognition
- speech recognition systems
- training set
- keyword spotting
- speech recognition technology
- noisy environments
- speaker identification
- speech understanding
- speech recognizers
- training process
- speech recognition errors
- n gram
- cepstral coefficients