On Spectral and Temporal Sparsification of Speech Signals for the Improvement of Speech Perception in CI Listeners.
Benjamin LentzRainer MartinKirsten OberländerChristiane VölterPublished in: ICASSP (2022)
Keyphrases
- speech signal
- linear prediction
- spectral analysis
- speech recognition
- spectral subtraction
- sound source
- automatic speech recognition
- automatic speech recognition systems
- vocal tract
- cepstral coefficients
- speaker identification
- computational intelligence
- background noise
- speech sounds
- noisy environments
- non stationary
- hidden markov models
- speech enhancement
- speaker recognition
- spectral features
- speech quality
- acoustic features
- noisy speech
- multiresolution
- speech synthesis
- noisy images
- language model
- denoising
- high resolution
- machine learning
- formant frequencies