Recognizing Overlapped Speech in Meetings: A Multichannel Separation Approach Using Neural Networks.
Takuya YoshiokaHakan ErdoganZhuo ChenXiong XiaoFil AllevaPublished in: INTERSPEECH (2018)
Keyphrases
- neural network
- audio visual
- speaker diarization
- linear prediction
- speech recognition
- pattern recognition
- speech signal
- automatic transcription
- fuzzy logic
- neural network model
- artificial neural networks
- cross channel
- neural nets
- speaker recognition
- activation function
- automatic speech recognition
- genetic algorithm
- text to speech
- fault diagnosis
- multi modal
- network architecture
- multi layer
- multi channel
- recurrent neural networks
- back propagation
- spoken language
- spoken dialogue systems
- recognition engine
- multilayer perceptron
- audio stream
- endpoint detection
- visual information