A Multi-level Acoustic Feature Extraction Framework for Transformer Based End-to-End Speech Recognition.
Jin LiRongfeng SuXurong XieLan WangNan YanPublished in: INTERSPEECH (2022)
Keyphrases
- end to end
- speech recognition
- feature extraction
- speech recognition systems
- pattern recognition
- speech signal
- hidden markov models
- speech recognizers
- speech processing
- speech synthesis
- cepstral coefficients
- speaker identification
- speech recognition technology
- congestion control
- automatic speech recognition
- language model
- probabilistic model
- machine learning
- speaker dependent
- isolated word
- speech recognizer
- noise reduction
- bitstream
- feature set
- feature selection