SiSEC 2015

NEWS: The evaluation results of SiSEC 2015 are now available.

NEWS: The submission deadline is extended to 25th May.

NEWS: We replaced the MUS dataset MSD100.zip to MSD100_2.zip because some songs in the older version were corrupted (the songs stop playing before the end). Please download them again from the MUS page.

Welcome to the main page for the fifth community-based Signal Separation Evaluation Campaign (SiSEC 2015).

SiSEC aims to be a large-scale regular campaign building upon the experience of previous evaluation campaigns and first community-based Signal Separation Evaluation Campaign (SASSEC).

So far, several SiSECs (SiSEC2008, SiSEC2010, SiSEC2011, and SiSEC2013) were held every one and half year. The unique aspect of this campaign is that it is not a competition but a scientific evaluation from which we can draw rigorous scientific conclusions.

Reference Papers

The results of this and previous evaluation campaigns are summarized in the following reference papers:

  • N. Ono, Z. Koldovsky, S. Miyabe, N. Ito, The 2013 Signal Separation Evaluation Campaign (SiSEC2013), in Proc. Int. Workshop on Machine Learning for Signal Processing, pp.1-6, 2013.
  • S. Araki, F. Nesta, E. Vincent, Z. Koldovsky, G. Nolte, A. Ziehe and A. Benichoux, The 2011 Signal Separation Evaluation Campaign (SiSEC2011): – Audio source separation -, in Proc. Int. Conf. on Latent Variable Analysis and Signal Separation, pp.414-422, 2012.
  • G. Nolte, D. Lutter, A. Ziehe, F. Nesta, E. Vincent, Z. Koldovsky, A. Benichoux and S. Araki, The 2011 Signal Separation Evaluation Campaign (SiSEC2011): – Biomedical data analysis -, in Proc. Int. Conf. on Latent Variable Analysis and Signal Separation, pp.423-429, 2012.
  • E. Vincent, S. Araki, F.J. Theis, G. Nolte, P. Bofill, H. Sawada, A. Ozerov, B.V. Gowreesunker, D. Lutter and N.Q.K. Duong, The Signal Separation Evaluation Campaign (2007-2010): Achievements and remaining challenges, Signal Processing, 92, pp.1928-1936, 2012.

Datasets, Tasks, and Evaluation Procedures

  1. Underdetermined speech and music mixtures
  2. Two-channel mixtures of speech and real-world background noise
  3. Professionally produced music recordings
  4. Asynchronous recordings of speech mixtures

We welcome other tasks/datasets/evaluation criteria from participants. Anyone can comment or reply about this issue via the comment system at the bottom of the task page.

Evaluation Results of SiSEC 2015

Important Days

  • End of Jan., 2015:  Deadline for evaluation proposal
  • Feb., 2015:             Call for submission of separation results
  • 27th Mar., 2015:     Paper submission deadline of LVA/ICA 2015
  • 20th May, 2015:  Deadline for submission of separation results
  • 25th May, 2015:  Deadline for submission of separation results (extended)
  • Beg. of Jun., 2015: Publication of SiSEC 2015 results (on the web)
  • 12th Jun., 2015:     Deadline for LVA/ICA camera-ready papers
  • 26-28 Aug., 2015:  LVA/ICA conference

SiSEC 2015 Evaluation Organizing Committee

  • Nobutaka Ono (NII)
  • Antoine Liutkus (INRIA-Nancy)
  • Daichi Kitamura (SOKENDAI)
  • Nobutaka Ito (NTT)
  • Zafar Rafii (Gracenote)

UND 2015

Underdetermined-speech and music mixtures We propose to repeat the underdetermined-speech and music mixtures task in SiSEC2013. Test data We have three datasets: Download test.zip (22 MB) (test data of SiSEC2008.) Download test2.zip (16 MB) (test data of SiSEC2010.) Download test3.zip (8.6MB)(test data of SiSEC2011. This is the 3-ch mixtures of 4 speech sources.) test.zip test.zip …

BGN 2015

Two-channel mixtures of speech and real-world background noise We propose to repeat the Two-channel mixtures of speech and real-world background noise without Chime corpus because the reference speech data has been already provided in the second ChiME challenge. Introduction This task aims to evaluate denoising and DOA estimation techniques by the SiSEC 2010 noisy speech …

MUS 2015

Professionally-produced music recordings 0. Introduction The purpose of this task is to evaluate source separation algorithms for estimating one or more sources from a set of mixtures in the context of professionally-produced music recordings. The data set consists of a total of 100 full-track songs of different styles and includes the synthesized mixtures and the …

ASY 2015

  Asynchronous recordings of speech mixtures Introduction Asynchronous recording is a new interesting task of source separation. The multichannel observation is obtained by multiple independent recording devices, and has wide range of application using portable recording devices as smartphones and IC recorders. The main differences from the conventional synchronous multichannel recording are unknown time offset …