A Speech Reconstruction Algorithm via Iteratively Reweighted ℓ2 Minimization for MFCC Codec.
Gang MinXiongwei ZhangXiangyang LiuChangqing ZhangYanpu ChenPublished in: EUSIPCO (2019)
Keyphrases
- speech recognition
- speech signal
- speaker identification
- speaker recognition
- automatic speech recognition
- speech recognizer
- mel frequency cepstral coefficients
- speech synthesis
- cepstral coefficients
- vocal tract
- background noise
- speech recognition systems
- noisy speech
- spoken language
- video coding
- acoustic features
- gaussian mixture model
- language model
- hidden markov models
- feature extraction
- spectral analysis
- objective function
- noisy environments
- audio visual
- linear prediction
- video codec
- coding method
- feature set
- audio features
- emotion recognition
- low bit rate
- non stationary
- recognition engine
- speech enhancement
- principal component analysis
- computational complexity
- hearing impaired
- linear predictive