Predicting algorithm efficacy for adaptive multi-cue source separation

Ethan Manilow, Prem Seetharaman, Fatemeh Pishdadian, Bryan A Pardo

Research output: Chapter in Book/Report/Conference proceedingConference contribution

11 Scopus citations

Abstract

Audio source separation is the process of decomposing a signal containing sounds from multiple sources into a set of signals, each from a single source. Source separation algorithms typically leverage assumptions about correlations between audio signal characteristics ('cues') and the audio sources or mixing parameters, and exploit these to do separation. We train a neural network to predict quality of source separation, as measured by Signal to Distortion Ratio, or SDR. We do this for three source separation algorithms, each leveraging a different cue-repetition, spatialization, and harmonicity/pitch proximity. Our model estimates separation quality using only the original audio mixture and separated source output by an algorithm. These estimates are reliable enough to be used to guide switching between algorithms as cues vary. Our approach for separation quality prediction can be generalized to arbitrary source separation algorithms.

Original languageEnglish (US)
Title of host publication2017 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, WASPAA 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages274-278
Number of pages5
ISBN (Electronic)9781538616321
DOIs
StatePublished - Dec 7 2017
Event2017 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, WASPAA 2017 - New Paltz, United States
Duration: Oct 15 2017Oct 18 2017

Publication series

NameIEEE Workshop on Applications of Signal Processing to Audio and Acoustics
Volume2017-October

Other

Other2017 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, WASPAA 2017
Country/TerritoryUnited States
CityNew Paltz
Period10/15/1710/18/17

Funding

∗contributed equally †This work was supported by NSF Grant 1420971.

Keywords

  • background
  • foreground
  • melody
  • prediction
  • repetition
  • singing voice separation
  • spatial

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Computer Science Applications

Fingerprint

Dive into the research topics of 'Predicting algorithm efficacy for adaptive multi-cue source separation'. Together they form a unique fingerprint.

Cite this