Multichannel signal separation for cocktail party speech recognition: a dynamic recurrent network.
Seungjin ChoiHeonseok HongHervé GlotinFrédéric BerthommierPublished in: Neurocomputing (2002)
Keyphrases
- speech recognition
- recurrent networks
- noisy environments
- hidden markov models
- automatic speech recognition
- language model
- speech recognizer
- pattern recognition
- speech synthesis
- speech processing
- speech signal
- speaker identification
- recurrent neural networks
- speech recognition systems
- speech recognition errors
- noisy speech
- isolated word
- single channel
- keyword spotting
- frequency domain
- neural network
- sound source
- speech retrieval
- signal processing
- speaker adaptation
- speech recognizers