Speech perception using a two-formant 22-electrode cochlear prosthesis in quiet and in noise

Area of Science:

• Audiology
• Biomedical Engineering
• Speech Processing

Background:

• Cochlear implants enhance hearing for individuals with severe to profound sensorineural hearing loss.
• Current cochlear implant speech processing strategies encode fundamental frequency (F0), voice pitch, and second formant (F2) frequencies.
• Limitations exist in speech perception, especially in noisy conditions, necessitating advanced processing strategies.

Purpose of the Study:

• To evaluate a novel speech processing strategy (F0F1F2) for cochlear implants.
• To compare the efficacy of the new F0F1F2 strategy against the existing F0F2 strategy.
• To assess speech recognition performance in quiet and noisy environments.

Main Methods:

• Two groups of cochlear implant patients (n=13 for F0F2, n=9 for F0F1F2) were tested 3 months post-surgery.
• Both groups received similar postoperative auditory training.
• Speech recognition was assessed using open-set and closed-set tests, including conditions with competing speech babble.

Main Results:

• The F0F1F2 group demonstrated significantly improved speech recognition, particularly in unaided listening conditions.
• In the presence of speech babble, the F0F2 strategy showed significant performance degradation at a 10 dB signal-to-noise ratio.
• Conversely, the F0F1F2 strategy maintained performance in noise, showing no significant degradation at the same signal-to-noise ratio.

Conclusions:

• The F0F1F2 speech processing strategy offers superior performance for cochlear implant users compared to the F0F2 strategy.
• This new strategy enhances speech understanding in challenging listening environments, such as those with background noise.
• The findings support the clinical adoption of the F0F1F2 strategy for improved cochlear implant outcomes.