Created by: Chitralekha Gupta
Affiliation: NUS, Singapore
Last edited on: 12th June 2018
Last edited by: Chitralekha Gupta
This is the code base for automatically obtaining aligned lyrics for solo-singing in Smule's Sing! karaoke DAMP dataset. It provides a cleaner annotated dataset.
Please refer to the following paper for details:
Gupta, C., Tong R., Li, H. and Wang, Y., 2018, September, "SEMI-SUPERVISED LYRICS AND SOLO-SINGING ALIGNMENT". Accepted for ISMIR 2018, Paris.(http://ismir2018.ircam.fr/doc/pdfs/30_Paper.pdf)
- This program is designed for monophonic (without background music) audio files.
- The python scripts main.py is the main file.
- Start with main.py, and follow the instructions in the file header.
- The script currently runs for one example file present in the folder "audio"
- You can add more files from the DAMP dataset to the "audio" folder
The code and models in this repository are licensed under the GNU General Public License Version 3. For commercial use of this code and models, separate commercial licensing is also available. Please see the contacts below.
- Chitralekha Gupta: chitralekha[at]u[dot]nus[dot]edu
- Haizhou Li: haizhou[dot]li[at]nus[dot]edu[dot]sg
- Ye Wang: wangye[at]comp[dot]nus[dot]edu[dot]sg