Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Enhancing Cochlear Implant Signal Coding with Scaled Dot-Product Attention

Billel Essaid, Hamza Kheddar, Noureddine Batel

TL;DR

This work addresses the limited adaptability of traditional cochlear implant coding by applying a deep-learning model that combines a temporal convolutional network with scaled dot-product attention to generate electrodograms. The approach leverages an encoder–TCN–attention architecture to capture both local and global temporal patterns in speech for CI stimulation, comparing its performance to ACE using the STOI metric. Results show the model achieves a STOI of 0.6031, closely matching ACE's 0.6126, demonstrating competitive intelligibility alongside greater flexibility. The work highlights AI-driven CI coding as a pathway to personalized and efficient stimulation strategies, with potential for real-time adaptation and cross-environment robustness.

Abstract

Cochlear implants (CIs) play a vital role in restoring hearing for individuals with severe to profound sensorineural hearing loss by directly stimulating the auditory nerve with electrical signals. While traditional coding strategies, such as the advanced combination encoder (ACE), have proven effective, they are constrained by their adaptability and precision. This paper investigates the use of deep learning (DL) techniques to generate electrodograms for CIs, presenting our model as an advanced alternative. We compared the performance of our model with the ACE strategy by evaluating the intelligibility of reconstructed audio signals using the short-time objective intelligibility (STOI) metric. The results indicate that our model achieves a STOI score of 0.6031, closely approximating the 0.6126 score of the ACE strategy, and offers potential advantages in flexibility and adaptability. This study underscores the benefits of incorporating artificial intelligent (AI) into CI technology, such as enhanced personalization and efficiency.

Enhancing Cochlear Implant Signal Coding with Scaled Dot-Product Attention

TL;DR

This work addresses the limited adaptability of traditional cochlear implant coding by applying a deep-learning model that combines a temporal convolutional network with scaled dot-product attention to generate electrodograms. The approach leverages an encoder–TCN–attention architecture to capture both local and global temporal patterns in speech for CI stimulation, comparing its performance to ACE using the STOI metric. Results show the model achieves a STOI of 0.6031, closely matching ACE's 0.6126, demonstrating competitive intelligibility alongside greater flexibility. The work highlights AI-driven CI coding as a pathway to personalized and efficient stimulation strategies, with potential for real-time adaptation and cross-environment robustness.

Abstract

Cochlear implants (CIs) play a vital role in restoring hearing for individuals with severe to profound sensorineural hearing loss by directly stimulating the auditory nerve with electrical signals. While traditional coding strategies, such as the advanced combination encoder (ACE), have proven effective, they are constrained by their adaptability and precision. This paper investigates the use of deep learning (DL) techniques to generate electrodograms for CIs, presenting our model as an advanced alternative. We compared the performance of our model with the ACE strategy by evaluating the intelligibility of reconstructed audio signals using the short-time objective intelligibility (STOI) metric. The results indicate that our model achieves a STOI score of 0.6031, closely approximating the 0.6126 score of the ACE strategy, and offers potential advantages in flexibility and adaptability. This study underscores the benefits of incorporating artificial intelligent (AI) into CI technology, such as enhanced personalization and efficiency.
Paper Structure (13 sections, 3 equations, 6 figures, 1 table)

This paper contains 13 sections, 3 equations, 6 figures, 1 table.

Table of Contents

  1. Introduction
  2. CIs Background
  3. CIs principle
  4. Stimulation strategies in CIs
  5. Enhancing CI coding strategies through AI
  6. AI and CI stimulation strategies
  7. TCN and scaled dot-product attention mechanisms in speech processing
  8. Proposed method
  9. DL model
  10. Audio material
  11. Objective metric
  12. Experimental results
  13. Conclusion

Figures (6)

  • Figure 1: Diagram of the components of a CI. brand2014cochlear.
  • Figure 2: Categories of signal processing strategies in CIs zeng2008cochlear.
  • Figure 3: Basic block diagram of ACE processing strategy hansen2019cci
  • Figure 4: Structure of the proposed model.
  • Figure 5: Training and validation performance
  • ...and 1 more figures