Deep Ad-hoc Beamforming Based on Speaker Extraction for Target-Dependent Speech Separation.
Ziye YangShanzheng GuanXiao-Lei ZhangPublished in: CoRR (2020)
Keyphrases
- speech recognition
- audio visual
- sound source
- speaker recognition
- automatic speech recognition
- speech signal
- speech enhancement
- speaker identification
- speaker verification
- vocal tract
- prosodic features
- noisy environments
- automatic speech recognition systems
- speaker diarization
- information extraction
- speech sounds
- noise reduction
- language model
- speech synthesis
- endpoint detection
- blind source separation
- emotion recognition
- linear prediction
- automatic extraction
- vector quantization
- speaker dependent
- hidden markov models
- acoustic features
- signal to noise ratio
- moving target
- automatic transcription
- neural network
- speech recognition systems
- probabilistic model
- linear array
- multi modal
- non stationary
- gaussian mixture model
- deep learning
- visual information
- dialogue system