Multi-talker Speech Separation Based on Permutation Invariant Training and Beamforming.
Lu YinZiteng WangRisheng XiaJunfeng LiYonghong YanPublished in: INTERSPEECH (2018)
Keyphrases
- sound source
- speech signal
- hearing impaired
- audio visual
- speech recognition
- speech enhancement
- training process
- training examples
- test set
- endpoint detection
- affine invariant
- automatic speech recognition
- decision trees
- supervised learning
- recognition engine
- speech synthesis
- speaker identification
- broadcast news
- spoken language
- training data
- feature vectors
- neural network
- training algorithm
- feature points