RESUMEN

It is well-established that hearing loss does not only lead to a reduction of hearing sensitivity. Large individual differences are typically observed among listeners with hearing impairment in a wide range of suprathreshold auditory measures. In many cases, audiometric thresholds cannot fully account for such individual differences, which make it challenging to find adequate compensation strategies in hearing devices. How to characterize, model, and compensate for individual hearing loss were the main topics of the fifth International Symposium on Auditory and Audiological Research (ISAAR), held in Nyborg, Denmark, in August 2015. The following collection of papers results from some of the work that was presented and discussed at the symposium.

Asunto(s)

RESUMEN

Object binding cues such as binaural and across-frequency modulation cues are likely to be used by the auditory system to separate sounds from different sources in complex auditory scenes. The present study investigates the interaction of these cues in a binaural masking pattern paradigm where a sinusoidal target is masked by a narrowband noise. It was hypothesised that beating between signal and masker may contribute to signal detection when signal and masker do not spectrally overlap but that this cue could not be used in combination with interaural cues. To test this hypothesis an additional sinusoidal interferer was added to the noise masker with a lower frequency than the noise whereas the target had a higher frequency than the noise. Thresholds increase when the interferer is added. This effect is largest when the spectral interferer-masker and masker-target distances are equal. The result supports the hypothesis that modulation cues contribute to signal detection in the classical masking paradigm and that these are analysed with modulation bandpass filters. A monaural model including an across-frequency modulation process is presented that account for this effect. Interestingly, the interferer also affects dichotic thresholds indicating that modulation cues also play a role in binaural processing.

Asunto(s)

RESUMEN

Acoustic-Phonetic features necessary for rule-based speech recognition are described. The selection of feature algorithms is based on their robustness to variability of dynamics in speech signals and to influence of environmental noise, and their suitability for real-time implementation using speech production or speech perception modelling. Features are estimated in real-time, using a 32-bit floating point signal processor DSP32. The features described are pitch, formants, segmentation and labelling. The paper indicates the potential use of these features in connection with future experiments leading to development of new hearing aids.