Singing Voice Separation Demonstration
Drag and drop an audio file here.
- Accepts .wav file only.
- The uploaded audio file is converted to mono audio (if it is stereo) and resampled to 16kHz.
- To save computational resource and data transfer, the server only computes the separation results of about 30sec from the beginning if the uploaded audio is longer than that size.
- Therefore we suggest you modify your audio data and provide a shorter audio snippet of the part you want to hear the separation result.
- The server won't keep the uploaded audio file. It will be deleted when the separation process ends.
- See the Github repository and my Qiita post (in Japanese) for details about the separation model!
- Contact me at Twitter or email : yimingwu11[at]fudan.edu.cn