Acoustic Modeling with Densely Connected Residual Network for Multichannel Speech Recognition.
Jian TangYan SongLirong DaiIan McLoughlinPublished in: INTERSPEECH (2018)
Keyphrases
- speech recognition
- speech recognition systems
- densely connected
- speaker independent
- speech recognizers
- sparsely connected
- hidden markov models
- speech recognizer
- automatic speech recognition
- language model
- acoustic models
- community structure
- noisy environments
- information networks
- pattern recognition
- speech processing
- speech signal
- speech synthesis
- speech recognition technology
- noisy speech
- network structure
- complex networks
- speaker identification
- community detection
- speech retrieval
- multi channel
- speaker adaptation
- feature vectors
- speaker dependent
- image processing