Separation of multi-speaker speech from an underdetermined mixture of sources.
Khaoula MraihiMohamed Anouar Ben MessaoudPublished in: SSD (2023)
Keyphrases
- blind separation
- source signals
- blind source separation
- speech recognition
- speech signal
- speaker recognition
- independent component analysis
- automatic speech recognition
- audio visual
- mixing matrix
- speaker identification
- source separation
- speaker verification
- vocal tract
- prosodic features
- speaker diarization
- speaker dependent
- acoustic features
- automatic speech recognition systems
- nonlinear functions
- independent components
- speech synthesis
- gaussian mixture model
- audio features
- audio stream
- text to speech
- information sources
- speech recognizer
- synthesized speech
- speech sounds
- language model
- mel frequency cepstral coefficients
- mixture model
- vector quantization
- gaussian distribution
- single channel
- non stationary
- multiple sources
- multi source
- emotion recognition
- phoneme recognition
- blind deconvolution
- spoken language
- hidden markov models