Combination and Comparison of Sound Coding Strategies Using Cochlear Implant Simulation With Mandarin Speech

Area of Science:

• Audiology
• Speech Processing
• Biomedical Engineering

Background:

• Cochlear implant (CI) sound coding strategies are crucial for restoring speech intelligibility.
• Optimizing CI strategies, especially for tonal languages like Mandarin, remains an active area of research.
• Combining existing strategies offers a potential avenue for enhancing CI performance.

Purpose of the Study:

• To evaluate the speech intelligibility of novel combined cochlear implant sound coding strategies using Mandarin sentences.
• To compare the performance of individual and combined strategies, including biologically-inspired hearing aid algorithm (BioAid), envelope enhancement (EE), and fundamental frequency modulation (F0mod), with the advanced combination encoder (ACE).

Main Methods:

• Developed four singular and four combinational CI sound coding strategies.
• Processed Mandarin sentences with speech-shaped noise using these strategies.
• Evaluated speech intelligibility using short-time objective intelligibility (STOI) and subjective sentence recognition tests with normal-hearing listeners.

Main Results:

• The envelope enhancement (EE) strategy slightly outperformed the advanced combination encoder (ACE) at signal-to-noise ratios of 5 dB and above.
• The combination of BioAid and EE (BioAid+EE) yielded the highest scores in both objective (STOI) and subjective speech intelligibility tests.
• No significant benefits were observed for BioAid, F0mod, or other combinational strategies in the CI simulation.

Conclusions:

• The BioAid+EE combination strategy demonstrates potential for improving speech intelligibility in cochlear implant users, particularly for Mandarin.
• Further research into combined coding strategies may lead to more effective CI sound processing.
• Findings provide valuable insights for future CI strategy design and Mandarin-specific research.