Raw Waveform Encoder with Multi-Scale Globally Attentive Locally Recurrent Networks for End-to-End Speech Recognition.
Max W. Y. LamJun WangChao WengDan SuDong YuPublished in: CoRR (2021)
Keyphrases
- speech recognition
- end to end
- recurrent networks
- multiscale
- recurrent neural networks
- rate allocation
- feed forward
- hidden markov models
- biologically inspired
- neural network
- pattern recognition
- speech signal
- rate distortion
- speech recognizer
- language model
- speech recognition technology
- automatic speech recognition
- speech synthesis
- bit rate
- congestion control
- frequency domain
- wavelet transform
- low complexity
- distributed video coding
- video compression
- video codec
- motion estimation
- image quality
- speaker identification
- artificial neural networks
- image processing
- speech recognition systems
- genetic algorithm