Role of masker predictability in the cocktail party problem
- 1 December 2008
- journal article
- research article
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 124 (6) , 3818-3830
Abstract
In studies of the cocktail party problem, the number and locations of maskers are typically fixed throughout a block of trials, which leaves out uncertainty that exists in real-world environments. The current experiments examined whether there is (1) improved speech intelligibility and (2) increased spatial release from masking (SRM), as predictability of the number/locations of speech maskers is increased. In the first experiment, subjects identified a target word presented at a fixed level in the presence of 0, 1, or 2 maskers as predictability of the masker configuration ranged from 10% to 80%. The second experiment examined speech reception thresholds and SRM as (a) predictability of the masker configuration is increased from 20% to 80% and/or (b) the complexity of the listening environment is decreased. In the third experiment, predictability of the masker configuration was increased from 20% up to 100% while minimizing the onset delay between maskers and the target. All experiments showed no effect of predictability of the masker configuration on speech intelligibility or SRM. These results suggest that knowing the number and location(s) of maskers may not necessarily contribute significantly to solving the cocktail party problem, at least not when the location of the target is known.Keywords
This publication has 44 references indexed in Scilit:
- Effect of spatial uncertainty of masker on masked detection for nonspeech stimuliThe Journal of the Acoustical Society of America, 2008
- Speech detection in spatial and nonspatial speech maskersThe Journal of the Acoustical Society of America, 2008
- Speech recognition with varying numbers and types of competing talkers by normal-hearing, cochlear-implant, and implant simulation subjectsThe Journal of the Acoustical Society of America, 2008
- Factors That Influence Intelligibility in Multitalker Speech DisplaysThe International Journal of Aviation Psychology, 2004
- Note on informational masking (L)The Journal of the Acoustical Society of America, 2003
- Release from masking due to spatial separation of sources in the identification of nonspeech auditory patternsThe Journal of the Acoustical Society of America, 1998
- A simulated “cocktail party” with up to three sound sourcesPerception & Psychophysics, 1996
- A model of auditory pattern analysis based on component-relative-entropyThe Journal of the Acoustical Society of America, 1993
- Perceptual Masking in Multiple Sound BackgroundsThe Journal of the Acoustical Society of America, 1969
- Some Experiments on the Recognition of Speech, with One and with Two EarsThe Journal of the Acoustical Society of America, 1953