Speech Enhancement by Reconstruction from Cleaned Acoustic Features.
Philip HardingBen MilnerPublished in: INTERSPEECH (2011)
Keyphrases
- speech signal
- acoustic features
- speech enhancement
- noisy environments
- speech recognition
- speaker verification
- automatic speech recognition
- vocal tract
- linear prediction
- non stationary
- speaker identification
- signal to noise ratio
- spectral analysis
- noise reduction
- hidden markov models
- speaker recognition
- high resolution
- single channel
- cross correlation
- visual features
- language model
- mel frequency cepstral coefficients
- noisy images
- audio features
- face recognition
- audio visual