Combined spectral and temporal enhancement to improve cochlear-implant speech perception

Abstract

The present study examined the effect of combined spectral and temporal enhancement on speech recognition by cochlear-implant (CI) users in quiet and in noise. The spectral enhancement was achieved by expanding the short-term Fourier amplitudes in the input signal. Additionally, a variation of the Transient Emphasis Spectral Maxima (TESM) strategy was applied to enhance the short-duration consonant cues that are otherwise suppressed when processed with spectral expansion. Nine CI users were tested on phoneme recognition tasks and ten CI users were tested on sentence recognition tasks both in quiet and in steady, speech-spectrum-shaped noise. Vowel and consonant recognition in noise were significantly improved with spectral expansion combined with TESM. Sentence recognition improved with both spectral expansion and spectral expansion combined with TESM. The amount of improvement varied with individual CI users. Overall the present results suggest that customized processing is needed to optimize performance according to not only individual users but also listening conditions.

Evaluation of near-end speech enhancement under equal-loudness constraint for listeners with normal-hearing and mild-to-moderate hearing loss

“…Some approaches are based on fairly basic forms of signal processing such as high-pass filtering followed by amplitude compression (Niederjohn and Grotelueschen, 1976), or signal-to-noise ratio (SNR) recovery (Sauert and Vary, 2006). Spectral contrast enhancement has been used as a possible method of compensating for the reduced frequency selectivity of hearing-impaired people (Baer et al, 1993;Oxenham et al, 2007;Bhattacharya et al, 2011).…”

The potential of onset enhancement for increased speech intelligibility in auditory prostheses

“…While the TESM strategy showed a significant improvement of 11.3% at an SNR of 5 dB in multitalker babble noise (Vandali, 2001), the same strategy did not differ significantly in a sentence recognition task in Holden et al (2005) when the target signal level was presented at 65 dB SPL. Bhattacharya et al (2011) even observed a decrease in speech intelligibility in quiet for the strategy and obtained a significant improvement when adding an additional spectral extension stage to the processing. The TESM strategy is more focused on the transient part and not the onset of the speech envelope.…”

“…Vandali (2001) developed the transient emphasis spectral maxima (TESM) strategy that amplifies transient parts of the signal with a gain factor derived from a comparison of the averaged energy in three consecutive time windows in each processing channel. In several studies with CI listeners ( Vandali, 2001;Bhattacharya et al, 2011), it was shown that small but significant improvements in speech intelligibility were obtained with the TESM strategy in quiet for consonant recognition and in multitalker babble noise at 5 dB SNR (Vandali, 2001) or in combination with a spectral expansion stage (Bhattacharya et al, 2011). In contrast, Holden et al (2005) found no increase in speech intelligibility with this strategy.…”

Auditory Perception and Production of Speech Feature Contrasts by Pediatric Implant Users

“…In contrast to vowels, the acoustic information associated with consonants tends to be more rapidly changing and related to formant transitions and other acoustic characteristics that would require fairly good auditory resolution of rapidly changing spectral information. Research exploring the relationship between the ability to perceive temporal and spectral modulation and speech perception suggests that these abilities are related and that limited modulation resolution may play a role in speech perception (Bhattacharya et al 2011;Won et al 2011), particularly for consonant perception (Van Tasell et al 1987).…”