An Empirical Study on Transformer-Based End-to-End Speech Recognition with Novel Decoder Masking.
Shi-Yan WengHsuan-Sheng ChiuBerlin ChenPublished in: APSIPA ASC (2021)
Keyphrases
- speech recognition
- end to end
- rate allocation
- hidden markov models
- language model
- human visual system
- speech recognizer
- noisy environments
- pattern recognition
- speech synthesis
- speech signal
- speech recognition technology
- error concealment
- automatic speech recognition
- speech recognition systems
- speaker identification
- low complexity
- visual quality
- congestion control
- error resilient
- computer vision
- neural network
- scalable video
- error control
- image quality
- speaker adaptation