Spectral tilt modelling with extrapolated GMMs for intelligibility enhancement of narrowband telephone speech.
Emma JokinenUlpu RemesMarko TakanenKalle J. PalomäkiMikko KurimoPaavo AlkuPublished in: IWAENC (2014)
Keyphrases
- speech signal
- speech recognition
- spectral analysis
- linear prediction
- gaussian mixture model
- spectral features
- speaker recognition
- automatic speech recognition
- image enhancement
- linear predictive coding
- speaker identification
- hidden markov models
- spectral images
- speaker adaptation
- speech synthesis
- image processing
- vocal tract
- cepstral coefficients
- noisy environments
- text to speech
- audio visual
- non stationary
- pattern recognition
- multispectral images
- maximum likelihood
- image analysis
- speech sounds