Enhancing the performance of a GMM-based speaker identification system in a multi-microphone setup.
Andreas StergiouAristodemos PnevmatikakisLazaros C. PolymenakosPublished in: INTERSPEECH (2006)
Keyphrases
- speaker identification
- gaussian mixture model
- speech recognition
- speaker diarization
- speaker recognition
- speech signal
- noisy environments
- broadcast news
- feature extraction
- automatic speech recognition
- probabilistic neural network
- mixture model
- computer vision
- audio features
- prediction accuracy
- language model
- mel frequency cepstral coefficients