Regularized subspace-based acoustic multichannel equalization for speech dereverberation.
Ina KodrasiSimon DocloPublished in: EUSIPCO (2013)
Keyphrases
- speech sounds
- acoustic features
- speech recognition
- speech recognition systems
- linear prediction
- speech signal
- emotional speech
- speech recognizers
- acoustic models
- prosodic features
- automatic speech recognition
- subspace learning
- formant frequencies
- speaker independent
- high dimensional data
- low dimensional
- sound source
- emotion recognition
- acoustic signal
- speaker recognition
- speech synthesis
- vocal tract
- principal component analysis
- single channel
- least squares
- subspace clustering
- speech enhancement
- speech recognizer
- noisy speech
- dimensionality reduction
- hidden markov models
- cross channel
- high dimensional
- audio visual
- feature space
- bird species
- speaker verification
- speaker identification
- multi channel
- visual features
- feature extraction
- source localization
- speech segments
- audio features
- vowel phonemes