Stream-based speaker segmentation using speaker factors and eigenvoices.
Fabio CastaldoDaniele ColibroEmanuele DalmassoPietro LafaceClaudio VairPublished in: ICASSP (2008)
Keyphrases
- speaker verification
- speaker recognition
- speech recognition
- medical images
- segmentation method
- audio visual
- segmentation algorithm
- image segmentation
- speaker diarization
- level set
- automatic speech recognition
- data streams
- region growing
- factors that influence
- segmentation accuracy
- shape prior
- real time
- multiscale
- segmented images
- fully automatic
- medical imaging
- speaker identification
- hidden markov models
- face recognition
- speaker adaptation
- speaker dependent
- automatic transcription