Gated Recurrent Context: Softmax-free Attention for Online Encoder-Decoder Speech Recognition.
Hyeonseung LeeWoo Hyun KangSung Jun CheonHyeongju KimNam Soo KimPublished in: CoRR (2020)
Keyphrases
- speech recognition
- hidden markov models
- language model
- speech recognizer
- speech processing
- pattern recognition
- speech synthesis
- speech signal
- automatic speech recognition
- low complexity
- video codec
- decoding process
- speech recognition systems
- noisy environments
- speech recognition technology
- cepstral coefficients
- speech retrieval
- speaker dependent
- speaker independent
- speaker identification
- error control
- neural network
- motion estimation
- speaker recognition
- distributed video coding
- rate distortion
- speech recognition errors
- computational complexity
- audio visual speech recognition