BLSTM-Based Confidence Estimation for End-to-End Speech Recognition.
Atsunori OgawaNaohiro TawaraTakatomo KanoMarc DelcroixPublished in: ICASSP (2021)
Keyphrases
- end to end
- speech recognition
- hidden markov models
- speech signal
- speech synthesis
- language model
- pattern recognition
- speech recognition technology
- speech processing
- automatic speech recognition
- speech recognizer
- noisy environments
- speech recognition systems
- speaker independent
- congestion control
- speaker identification
- speech recognizers
- feature extraction
- background noise
- non stationary
- mobile devices
- speaker dependent
- information retrieval