End-to-end Audiovisual Speech Activity Detection with Bimodal Recurrent Neural Models.
Fei TaoCarlos BussoPublished in: CoRR (2018)
Keyphrases
- end to end
- neural models
- spiking neural networks
- smart room
- recurrent neural networks
- speaker diarization
- biologically inspired
- neural model
- feed forward
- neural network
- learning rules
- neural network model
- bio inspired
- multimedia content
- multi modal
- artificial neural networks
- audio visual
- biologically plausible
- visual information
- congestion control
- smart spaces
- motor control
- speech recognition
- word error rate
- training algorithm
- automatic speech recognition
- high dimensional
- web services
- multimedia