On combining DNN and GMM with unsupervised speaker adaptation for robust automatic speech recognition.
Shilin LiuKhe Chai SimPublished in: ICASSP (2014)
Keyphrases
- automatic speech recognition
- speaker adaptation
- speech recognition
- noisy environments
- speaker identification
- speech signal
- hidden markov models
- maximum likelihood
- conversational speech
- gaussian mixture model
- broadcast news
- speaker dependent
- speech retrieval
- speech recognizer
- semi supervised
- acoustic features
- speaker independent
- mixture model
- error rate
- feature vectors
- pattern recognition