TasNet: Surpassing Ideal Time-Frequency Masking for Speech Separation.
Yi LuoNima MesgaraniPublished in: CoRR (2018)
Keyphrases
- speech recognition
- speech synthesis
- human visual system
- signal processing
- speech signal
- automatic speech recognition
- endpoint detection
- recognition engine
- pattern recognition
- empirically derived
- speech processing
- spoken language
- frequency domain
- audio visual
- fourier transform
- speaker recognition
- text to speech
- wavelet transform
- linear prediction
- spoken dialogue systems
- signal analysis
- multimodal interfaces
- human computer interaction
- spontaneous speech
- hidden markov models
- audio stream
- computer vision