Speech Enhancement Using Forked Generative Adversarial Networks with Spectral Subtraction.
Ju LinSufeng NiuZice WeiXiang LanAdriaan J. van WijngaardenMelissa C. SmithKuang-Ching WangPublished in: INTERSPEECH (2019)
Keyphrases
- speech enhancement
- spectral subtraction
- hybrid method
- speech signal
- noisy environments
- signal to noise ratio
- noisy speech
- noise reduction
- single channel
- speech recognition
- linear prediction
- hearing aids
- background noise
- automatic speech recognition
- vocal tract
- smoothing algorithm
- edge preserving
- multi channel
- wavelet transform
- hidden markov models