Building Corpora for Single-Channel Speech Separation Across Multiple Domains.
Matthew MaciejewskiGregory SellLeibny Paola García-PereraShinji WatanabeSanjeev KhudanpurPublished in: CoRR (2018)
Keyphrases
- single channel
- multiple domains
- speech enhancement
- multi channel
- sound source
- cross domain
- speech signal
- speech recognition
- prior information
- frequency domain
- noise reduction
- natural language processing
- noisy environments
- independent component analysis
- wiener filter
- vocal tract
- signal to noise ratio
- data analysis
- wavelet decomposition
- heart rate