Effects of nonlinear frequency compression on Mandarin speech and sound-quality perception in hearing-aid users

Abstract

Objective: The purpose of the present study was to examine the effects of NLFC fitting in hearing aids and auditory acclimatisation on speech perception and sound-quality rating in hearing-impaired, native Mandarin-speaking adult listeners.

Design: Mandarin consonant, vowel and tone recognition were tested in quiet and sentence recognition in noise (speech-shaped noise at a +5 dB signal-to-noise ratio) with NLFC-on and NLFC-off. Sound-quality ratings were collected on a 0–10 scale at each test session. A generalised linear model and correlational analyses were performed.

Study sample: Thirty native Mandarin-speaking adults with moderate-to-severe sensorineural hearing loss were recruited.

Results: The hearing-impaired listeners showed significantly higher accuracy with NLFC-on than with NLFC-off for consonant and sentence recognition and the recognition performance improved with both NLFC-on and off as a function of increased length of use. The satisfaction score of sound-quality ratings for different types of sounds significantly increased with NLFC-on than with NLFC-off. The speech recognition results showed moderate to strong correlation with the unaided hearing thresholds.

Conclusion: For native Mandarin-speaking listeners with hearing loss, the NLFC technology provided modest but significant improvement in Mandarin fricative and sentence recognition. Subjectively, the naturalness and overall preference of sound-quality satisfaction judgement also improved with NLFC.

Keywords:

• Hearing aids
• speech perception
• nonlinear frequency compression
• sound quality

Acknowledgements

Solveig Voss, Yitao Mao, Beier Qi, Yu Zhang, Lexi Neltner and Lauren Muscari provided technique assistance in various stages of the research.

Disclosure statement

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

Additional information

Funding

This research was funded in part by a research contract from Sonova AG.