A scalable frequency domain-based linear convolution architecture for speech enhancement.
Christelle YemdjiMoctar Mossi IdrissaNicholas W. D. EvansChristophe BeaugeantPublished in: DSP (2011)
Keyphrases
- frequency domain
- speech enhancement
- single channel
- fourier domain
- fourier transform
- spatial domain
- discrete fourier transform
- fast fourier transform
- linear prediction
- power spectrum
- feature extraction
- subband
- noisy environments
- denoising
- filter bank
- noise reduction
- fir filters
- signal to noise ratio
- image compression
- additive noise
- transfer function
- multi channel
- wiener filter
- natural images
- vocal tract