Mandarin speech-in-noise and tone recognition using vocoder simulations of the temporal limits encoder for cochlear...

Area of Science:

• Auditory Neuroscience
• Signal Processing
• Biomedical Engineering

Background:

• Cochlear implant (CI) systems primarily rely on temporal envelope processing.
• Limitations in conveying temporal fine structure (TFS) hinder CI user performance.
• Effective TFS delivery strategies for CI remain an active research area.

Purpose of the Study:

• To introduce and theoretically analyze the temporal limits encoder (TLE) strategy for CI signal processing.
• To compare the performance of TLE against established (CIS) and experimental strategies.
• To evaluate the potential of TLE for enhancing TFS representation in CI.

Main Methods:

• Theoretical analysis comparing TLE with CIS and harmonic-single-sideband-encoder strategies.
• Perceptual experiments using vocoder simulations in normal-hearing listeners.
• Assessment of TLE and CIS performance on Mandarin speech reception in noise and tone recognition tasks.

Main Results:

• TLE modulator demonstrated superior or comparable performance to the CIS modulator in perceptual experiments.
• TLE showed advantages in representing both temporal envelope and slow-varying TFS.
• Performance varied slightly based on vocoder type (tone-band vs. noise-band).

Conclusions:

• The TLE strategy presents a viable approach for improving TFS representation in cochlear implants.
• This research suggests potential for enhanced auditory perception in CI users through advanced signal processing.
• TLE offers a promising avenue for future CI technology development.