Favourable and unfavourable short duration segments of speech in noise

Hu~mm perceptual cxperimcnts are described that present listeners with segmented stop consonant speech stimuli in noise. The selection of short duration speech scgmcnts is based on a local measure of the signal-to-noise ratio calculated over Ims windows. The aim is to create stimuli with known fluctuations occurring bctwcen a speech and noise sample to assess whether the presence of short duration “gaps” in the noise product favourable and unfavourable signal regions that influence identification. Perceptual results arc reported that suggest human listeners make better use of signals that comprise only 01’ positive, local signal-to-noise ratio segments. Such regions are assumed to be more favourdble I’or stimuli identification. Presenlation of stimuli containing only ncgativc signal-to-noise ratio regions does not appear to contribute as much. A model that is based on the accumulation of short duration spectral segments is presented that produces a similar set 01‘ identilication functions For the same test stimuli.