HTTP/2 301 server: GitHub.com content-type: text/html location: https://sagiebenaim.github.io/Singing/ x-github-request-id: 97A7:444BC:7D5CBA:8C77F3:695156E4 accept-ranges: bytes age: 0 date: Sun, 28 Dec 2025 16:12:20 GMT via: 1.1 varnish x-served-by: cache-bom-vanm7210064-BOM x-cache: MISS x-cache-hits: 0 x-timer: S1766938341.580548,VS0,VE197 vary: Accept-Encoding x-fastly-request-id: ee992e187b47aaf14491627b91f033a58f09aa95 content-length: 162 HTTP/2 200 server: GitHub.com content-type: text/html; charset=utf-8 last-modified: Thu, 09 May 2019 08:06:48 GMT access-control-allow-origin: * etag: W/"5cd3df98-3fb9" expires: Sun, 28 Dec 2025 16:22:20 GMT cache-control: max-age=600 content-encoding: gzip x-proxy-cache: MISS x-github-request-id: 380A:2DDCFF:7C9D3D:8BB640:695156E4 accept-ranges: bytes age: 0 date: Sun, 28 Dec 2025 16:12:21 GMT via: 1.1 varnish x-served-by: cache-bom-vanm7210064-BOM x-cache: MISS x-cache-hits: 0 x-timer: S1766938341.791324,VS0,VE220 vary: Accept-Encoding x-fastly-request-id: 7a34c3642be5dc7a3935430c388a45d3c6c17e5f content-length: 1948

Samples for Semi-supervised Monaural Singing Voice Separation with a Masking Network Trained on Synthetic Mixtures

A comparison of our semi-supervised method with the state of the art supervised method of "Monaural Singing Voice Separation with Skip-Filtering Connections and Recurrent Inference of Time-Frequency Mask" by Mimilakis et al., ICASSP 2018.

All baseline results are obtained from https://js-mim.github.io/mss_pytorch/.

It seems that our results are cleaner, in the sense that the singing contains less instrumental music. However, there are more distortions in our results. This is reflected in the somewhat lower SDR we display but the much higher SIR.

Audio Demo on Evaluation subset: (please wait for the audio files to load)

All other music samples used by Mimilakis et al., ICASSP 2018: (please wait for the audio files to load)