Improving the Intelligibility of Electric and Acoustic Stimulation Speech Using Fully Convolutional Networks Based Speech Enhancement.
Natalie Yu-Hsien WangHsiao-Lan Sharon WangTaowei WangSzu-Wei FuXugang LuYu TsaoHsin-Min WangPublished in: CoRR (2019)
Keyphrases
- speech enhancement
- signal to noise ratio
- noisy speech
- speech recognition
- speech signal
- noise reduction
- noisy environments
- vocal tract
- single channel
- hidden markov models
- acoustic features
- linear prediction
- wiener filter
- speech sounds
- background noise
- smoothing algorithm
- sound source
- speech synthesis
- speaker verification
- automatic speech recognition
- language model
- speaker identification
- additive noise
- edge preserving
- image acquisition
- feature vectors