Suppression of Late Reverberation Effect on Speech Signal Using Long-Term Multiple-step Linear Prediction.
Keisuke KinoshitaMarc DelcroixTomohiro NakataniMasato MiyoshiPublished in: IEEE Trans. Speech Audio Process. (2009)
Keyphrases
- linear prediction
- speech signal
- speech recognition
- image coding
- vocal tract
- speech enhancement
- lossless compression
- prediction error
- speaker identification
- noisy environments
- automatic speech recognition
- spectral analysis
- non stationary
- sound source
- hidden markov models
- speaker recognition
- cepstral coefficients
- autoregressive
- subband
- acoustic features
- wavelet transform
- multiresolution
- pattern recognition
- bayesian networks
- high quality
- speech quality
- neural network