Bottleneck feature-mediated DNN-based feature mapping for throat microphone speech recognition.
Takahito SuzukiJun OgataTakashi TsunakawaMasafumi NishidaMasafumi NishimuraPublished in: APSIPA (2018)
Keyphrases
- speech recognition
- feature mapping
- automatic speech recognition
- speaker diarization
- hidden markov models
- cepstral coefficients
- speech synthesis
- language model
- speech signal
- pattern recognition
- speaker identification
- speech recognizer
- noisy environments
- speech recognition technology
- feature vectors
- manifold learning
- feature set
- spam filtering
- feature space
- speech recognition systems
- speaker independent
- nonlinear dimensionality reduction
- image features
- classification accuracy
- multiple instance learning
- broadcast news
- similarity measure
- feature extraction