The linear prediction inverse modulation transfer function (LP-IMTF) filter for spectral enhancement, with applications to speaker recognition.
Bengt J. BorgstromAlan McCreePublished in: ICASSP (2012)
Keyphrases
- linear prediction
- speaker recognition
- speech signal
- speaker identification
- speech recognition
- gaussian mixture model
- spectral analysis
- image coding
- prediction error
- speaker verification
- vector quantization
- subband
- noisy environments
- automatic speech recognition
- hidden markov models
- lossless compression
- non stationary
- image processing
- probabilistic neural network
- image enhancement
- image compression
- autoregressive
- language model
- contrast enhancement
- transfer function
- denoising
- pairwise
- pattern recognition