6533b86efe1ef96bd12cb28f
RESEARCH PRODUCT
Improving Isolation of Blindly Separated Sources Using Time-Frequency Masking
Maximo CobosJose J. Lopezsubject
Masking (art)business.industryComputer scienceApplied MathematicsSpeech recognitionPattern recognitioncomputer.software_genreBlind signal separationIndependent component analysisTime–frequency analysisSignal ProcessingSource separationArtificial intelligenceIsolation (database systems)Electrical and Electronic EngineeringAudio signal processingbusinesscomputerEnergy (signal processing)description
A refinement technique based on time-frequency masking is proposed to improve source isolation in blind audio source separation algorithms. The refinement technique uses an energy-normalized source-to-interference ratio in order to identify and eliminate interfering energy from the extracted sources. Some examples using this refinement method with different separation algorithms are discussed. The results show that source isolation can be significantly enhanced with negligible degradation of the separated sources.
| year | journal | country | edition | language |
|---|---|---|---|---|
| 2008-01-01 | IEEE Signal Processing Letters |