Robust spectral representation using group delay function and stabilized weighted linear prediction for additive noise degradations.
Dhananjaya N. GowdaJouni PohjalainenPaavo AlkuMikko KurimoPublished in: SpeD (2013)
Keyphrases
- linear prediction
- additive noise
- speech signal
- speech enhancement
- image coding
- noisy environments
- speech recognition
- noisy images
- spectral analysis
- spatial domain
- vocal tract
- lossless compression
- prediction error
- maximum likelihood
- hidden markov models
- cepstral coefficients
- autoregressive
- random variables
- subband
- non stationary
- motion estimation
- machine learning
- language model