On the Use of Spectrogram Inversion for Speech Enhancement.
Raja Abdelmalek BedouiZied MnasriFaouzi BenzartiPublished in: SSD (2021)
Keyphrases
- speech enhancement
- speech signal
- single channel
- multi channel
- speech recognition
- noisy environments
- vocal tract
- noise reduction
- linear prediction
- automatic speech recognition
- signal to noise ratio
- non stationary
- background noise
- independent component analysis
- image reconstruction
- wiener filter
- frequency domain
- spectral analysis
- prior information
- wavelet decomposition
- speech synthesis
- sound source
- hidden markov models
- brain activity
- additive noise
- computer vision
- feature space