Perception of Temporal Fine Structure Speech and Recovered Envelope Speech in Younger and Older Adults with Normal Hearing Sensitivity

Abstract

Perception of temporal fine structure (TFS) speech and recovered envelope (RENV) speech were compared between seven younger adults and seven older adults with normal hearing sensitivity. To create TFS speech, Kannada sentences were processed to remove temporal envelope cues while retaining TFS cues in 2, 4 and 8 frequency bands spanning the range of 80 to 8020 Hz. To create RENV speech, envelope cues were recovered from TFS speech extracted from 2 frequency bands, by passing it through 40 band-pass filters. RENV speech was also generated with simulated widened auditory filters, 2 and 4 times the normal auditory filter bandwidth. The findings show a general trend of reduction in scores with increase in number of frequency bands in TFS speech. The scores were significantly different across conditions when the age groups are combined, whereas significant difference between age groups was seen only for TFS speech extracted from 4 bands. Similarly, the scores reduced with simulation of cochlear hearing loss for perception of RENV speech. The scores were significantly different across the RENV speech conditions in both age groups. No significant difference was seen between the two age groups in any of the RENV speech conditions. The results of the study indicate that the ability to perceive TFS cues for sentence identification would degrade significantly with increase in the severity of cochlear pathology, without any significant age effect.