Skip to main content

Table 5 Performance measures for the best-case run of the separation of B /C tin whistle and viola/violin, with spectral masking. Results indicated as “Orig.” were generated from the dictionary that was trained on that recording, while “Gen.” means that the dictionary was trained on the other recording

From: Sparse pursuit and dictionary learning for blind source separation in polyphonic music recordings

Mode Instrument SDR SIR SAR
Orig. Tin whistle B 15.0 29.3 15.1
  Viola 10.5 26.9 10.6
  Tin whistle C 17.1* 27.0 17.6*
  Violin 12.1* 36.4* 12.1*
Gen. Tin whistle B 15.9* 30.0* 16.1*
  Viola 11.2* 28.4* 11.3*
  Tin whistle C 16.7 27.3* 17.1
  Violin 11.6 34.5 11.6
  1. Best numbers are marked