A Hands-On Comparison of DNNs for Dialog Separation Using Transfer Learning from Music Source Separation.
Martin StraussJouni PaulusMatteo TorcoliBernd EdlerPublished in: Interspeech (2021)
Keyphrases
- transfer learning
- source separation
- audio features
- cross domain
- machine learning
- labeled data
- blind source separation
- reinforcement learning
- collaborative filtering
- semi supervised learning
- independent component analysis
- text categorization
- learning algorithm
- transfer knowledge
- activity recognition
- active learning
- machine learning algorithms
- text classification
- natural language
- text mining
- denoising
- semi supervised
- audio visual
- target domain
- music information retrieval
- sound source
- audio signal
- video sequences