Online blind speech separation using multiple acoustic speaker tracking and time-frequency masking.
Pasi PertiläPublished in: Comput. Speech Lang. (2013)
Keyphrases
- speech recognition
- real time
- speaker recognition
- speech sounds
- acoustic features
- speaker verification
- audio visual
- prosodic features
- automatic speech recognition
- speech signal
- speaker identification
- acoustic models
- particle filter
- speech recognition systems
- online learning
- speaker independent
- speaker diarization
- hidden markov models
- broadcast news
- kalman filter
- frequency domain
- multi modal
- text to speech
- multiple objects
- speech synthesis
- visual speech
- motion model
- mean shift
- wavelet transform
- automatic speech recognition systems