Auditory model inspired CNN-based machine hearing model on end-to-end singing voice separation
碩士 === 國立交通大學 === 電機工程學系 === 107 === There are many different approaches for singing voice separation on frequency domain, and the most common way is to use various deep learning models in spectrum to construct a mask for each source. This paper takes the auditory model as the starting point. In the...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Published: |
2018
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/sxm896 |