Abstract
Harmonic complex tones are a particularly important class of sounds found in both speech and music. Although these sounds contain multiple frequency components, they are usually perceived as a coherent whole, with a pitch corresponding to the fundamental frequency (F0). However, when two or more harmonic sounds occur concurrently, e.g., at a cocktail party or in a symphony, the auditory system must separate harmonics and assign them to their respective F0s so that a coherent and veridical representation of the different sounds sources is formed. Here we review both psychophysical and neurophysiological (single-unit and evoked-potential) findings, which provide some insight into how, and how well, the auditory system accomplishes this task. A survey of computational models designed to estimate multiple F0s and segregate concurrent sources is followed by a review of the empirical literature on the perception and neural coding of concurrent harmonic sounds, including vowels, as well as findings obtained using single complex tones with mistuned harmonics.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 36-51 |
| Number of pages | 16 |
| Journal | Hearing Research |
| Volume | 266 |
| Issue number | 1-2 |
| DOIs | |
| State | Published - Jul 2010 |
Bibliographical note
Funding Information:This work was supported by a grant from NIDCD ( R01 DC05216 ). The authors are indebted to Brian C.J. Moore for many helpful comments, and to one anonymous reviewer for constructive suggestions on an earlier version of this manuscript. The manuscript also benefited from comments from Claude Alain, Yonatan Fishman, Martin McKinney, Erik Larsen, Alan Palmer, and Donald Sinex, as well as from discussions with Alain de Cheveigné and Peter Cariani.
Keywords
- Auditory scene analysis
- Complex tones
- Double vowels
- Fundamental frequency
- Harmonicity
- Mistuned harmonic
- Pitch