Multichannel Loss Function for Supervised Speech Source Separation by Mask-Based Beamforming.
Yoshiki MasuyamaMasahito TogamiTatsuya KomatsuPublished in: INTERSPEECH (2019)
Keyphrases
- loss function
- source separation
- single channel
- blind source separation
- speech signal
- sound source
- audio features
- frequency domain
- multi channel
- speech recognition
- audio visual
- pairwise
- independent component analysis
- support vector
- non stationary
- semi supervised
- supervised learning
- hidden markov models
- learning algorithm
- prior information
- machine learning
- spectral clustering
- denoising
- negative matrix factorization
- information retrieval
- cross validation
- wiener filter
- brain activity