Interspeech'2005 - Eurospeech

Lisbon, Portugal
September 4-8, 2005

Sound Segregation Based on Binaural Zero-Crossings

Young-Ik Kim, Sung Jun An, Rhee Man Kil, Hyung-Min Park

KAIST, Korea

This paper presents a new method of sound segregation based on zero-crossings generated from binaural filter-bank outputs. In our approach, sound source directions are identified using the spatial cues such as inter-aural time differences (ITDs) and inter-aural intensity differences (IIDs). The estimation of ITDs is performed using zero-crossings generated from binaural filter-bank outputs to get more reliable ITD samples in noisy environments. We also consider the estimation of ITDs with the aid of IID samples to cope with the phase ambiguities of ITD samples in high frequencies. As a result, the proposed method is able to provide an accurate estimate of sound source directions which gives us a good masking scheme for sound segregation while offering significantly less computational complexity compared to cross-correlation based methods.

Full Paper

Bibliographic reference.  Kim, Young-Ik / An, Sung Jun / Kil, Rhee Man / Park, Hyung-Min (2005): "Sound segregation based on binaural zero-crossings", In INTERSPEECH-2005, 2325-2328.