Branchformer: Parallel MLP-Attention Architectures to Capture Local and Global Context for Speech Recognition and Understanding.
Yifan PengSiddharth DalmiaIan R. LaneShinji WatanabePublished in: CoRR (2022)
Keyphrases
- speech recognition
- global context
- language model
- global information
- speech recognizer
- hidden markov models
- noisy environments
- speech signal
- automatic speech recognition
- speech synthesis
- pattern recognition
- named entities
- speech recognition technology
- speech recognition systems
- neural network
- sift descriptors
- multilayer perceptron
- speaker identification
- speaker independent
- co occurrence
- keywords
- computer vision