Candidate Speech Extraction from Multi-speaker Single-Channel Audio Interviews.
Meghna PandharipandeSunil Kumar KopparapuPublished in: SPECOM (1) (2023)
Keyphrases
- single channel
- audio visual
- speech enhancement
- audio stream
- speaker identification
- prosodic features
- multi channel
- speaker verification
- automatic transcription
- speech recognition
- sound source
- speaker recognition
- acoustic features
- text to speech
- automatic speech recognition
- speech synthesis
- audio features
- multi modal
- speaker diarization
- emotion recognition
- speech signal
- broadcast news
- visual information
- vocal tract
- multimedia
- visual data
- noisy environments
- signal processing
- frequency domain
- mel frequency cepstral coefficients
- prior information
- visual speech
- independent component analysis
- wiener filter
- spontaneous speech
- gaussian mixture model
- signal to noise ratio
- brain activity
- synthetic and real images
- prior knowledge