Speaker2Vec: Unsupervised Learning and Adaptation of a Speaker Manifold Using Deep Neural Networks with an Evaluation on Speaker Segmentation.
Arindam JatiPanayiotis G. GeorgiouPublished in: INTERSPEECH (2017)
Keyphrases
- neural network
- unsupervised learning
- audio visual
- speaker verification
- speech recognition
- speaker recognition
- pattern recognition
- speaker adaptation
- supervised learning
- prosodic features
- level set
- segmentation algorithm
- automatic speech recognition
- speaker diarization
- speaker identification
- synthesized speech
- deep learning
- object recognition
- image sequences
- image segmentation