1 / 33

Hearing loss and sparse coding

Hearing loss and sparse coding. Stefan Bleeck, Institute of Sound and Vibration Research, Hearing and Balance Centre University of Southampton . Question:. Can sparse coding help to overcome problems caused by hearing loss? overview of the hearing process

emmet
Download Presentation

Hearing loss and sparse coding

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Hearing loss and sparse coding Stefan Bleeck, Institute of Sound and Vibration Research, Hearing and Balance Centre University of Southampton

  2. Question: • Can sparse coding help to overcome problems caused by hearing loss? • overview of the hearing process • Examples of sparse algorithms for hearing aids and cochlear implants • Preliminary results bleeck@gmail.com

  3. a short journey into Hearing bleeck@gmail.com

  4. The outer ear Important for sound localization, linear => boring bleeck@gmail.com

  5. tympanum (middle ear) Important to explain limits of hearing, linear => boring bleeck@gmail.com

  6. the inner ear and the vestibular system • Contained within bony labyrinth in temporal bone • Cochlea does hearing • Semicircularcanals+utricle does balance • Same mechanism, nerve, evolution, similar problems bleeck@gmail.com

  7. frequency mapping bleeck@gmail.com 7

  8. Stereocilia • Stereocilia detect vibrations within cochlea. • Introduce half-wave rectification • Nonlinear bleeck@gmail.com 8

  9. range of loudness 10 1 10-2 10-4 13 orders of magnitude 10-6 10-8 10-10 10-12 bleeck@gmail.com

  10. ABSOLUTE THRESHOLD CURVE Threshold as function of Frequency membranemoves 10-13 m bleeck@gmail.com

  11. Cochlear nonlinearities • Amplitude (nonlinear amplification) • Frequencies (combination tones) • compression Demo: sweeps bleeck@gmail.com 11

  12. Active Processes healthy “active” OHCs provide up to 40 dB amplification(= factor of 100) Travelling Wave Envelope on Basilar Membrane due to Pure Tone Stimulus: damaged “passive” BM displacement (nm) Distance along BM (mm) OHCs inject energy in this region

  13. hearing loss (noise induced and presbyacousis) • 50% of 60 year old, 90% of 80 year old • Hearing aids are not good enough • ‘damage’ €2.4 Billion per year in EU • Lack of research funding today bleeck@gmail.com 13

  14. Hair cell loss by noise exposure Electron micrographs of cochlear hair cells. Left: healthy, right: damaged by noise exposure. bleeck@gmail.com 14

  15. Normal and impairedAuditory filter shapes • Hearing impairment • loosing audibility, • Also widening of filter • both results in difficulties to understand language, especially in noise bleeck@gmail.com 15

  16. The challenge

  17. Listening in noise Hearing Impaired 100% Aided Un-aided Normal Word recognition ASR 50% 0% 0 dB -15 dB 40 dB SNR

  18. bleeck@gmail.com

  19. ‘standard’ methods of denoising bleeck@gmail.com

  20. What we think a better solution could be: • Problem: Hearing loss constitutes a bottle neck: not all information can get through • Solution: extract less, but important information • Extract content based on Information not on Energy • Specifically speech related information bleeck@gmail.com

  21. Bio-inspired approach to denoising speech 3 Denoising (sparsification) 2 Neural representation: (Transformation) 1 periphery model

  22. ‘Sparse’ algorithms developed in our group noise bleeck@gmail.com

  23. Filter bank bleeck@gmail.com

  24. Example: nmf • Non-negative matrix factorization • Matrix Z is factorised into two non-negative matrices W and H (basis vectors (5) and activity over time) • (motivated by the processing in CI and auditory neurons) • Z here is the ‘envelopegram’ (22 channels, 128 pt) • Factorization using Euclidean cost function: • Sparseness constrained: g(H)= regularity function λ= sparsity factor bleeck@gmail.com

  25. Iterative algorithm to minimize the cost function by gradient decent: λ depends on SNR because of trade-off intelligibility - quality low noise: no sparsification high noise: lots Task: fine out how! Online experiment (restricted by speed of hardware) Offline experiment (unrestricted) bleeck@gmail.com

  26. For ‘bin’, ‘pin’, ‘din’, ‘tin’ W Z H bleeck@gmail.com

  27. bleeck@gmail.com

  28. On-line experimental set up: bleeck@gmail.com

  29. Sound examples: • 22 channel filter bank • 16 msframes • Gaussian noise SNR=5 dB clean noisy frequency denoised time

  30. Results from CI listeners in online experiment (problems with iteration!) bleeck@gmail.com

  31. Results from CI listeners in offline experiment results for all participants Best sparsification as function of snr: Averaged bleeck@gmail.com

  32. Conclusions: • Sparse coding can help reduce acoustic information in a useful way • Development still in its infancy, hardware restrictions still relevant • High impact research field with lots of potential funding • Strength of our group: clinical evaluation, • weakness at the moment: lack of signal processing experts

  33. Hu, H., Li, G., Chen, L., Sang, J., Wang, S., Lutman, M. E., & Bleeck, S. (2011). Enhanced sparse speech coding strategy for cochlear implants. European Signal Processing Conference (EUSIPCO). • Hu, H., Taghia, J., Sang, J., Taghia, J., Mohammadiha, N., Azarpour, M., Dokku, R., et al. (2011). Speech Enhancement via Combination of Wiener Filter and Blind Source Separation. International Conference on Intelligent Systems and Knowledge Engineering. • Sang, J., Hu, H., Li, G., Lutman, M. E., & Bleeck, S. (2011a). Application of a sparse coding strategy to enhance speech perception for hearing aid users. British Society of Audiology Short Papers Meeting. • Sang, J., Hu, H., Li, G., Lutman, M. E., & Bleeck, S. (2011b). Enhanced Sparse Speech Processing Strategy in Cochlear Implants. Conference on implantable Auditory Prostheses (CIAP). • Sang, J., Li, G., Hu, H., Lutman, M. E., & Bleeck, S. (2011a). Supervised Sparse Coding in Cochlear Implants. Conference on implantable Auditory Prostheses (CIAP). • Sang, J., Li, G., Hu, H., Lutman, M. E., & Bleeck, S. (2011b). Supervised Sparse Coding Strategy in Hearing Aids. Annual Conference of the International Speech Communication Association (INTERSPEECH). • Bleeck, S., Wright, M. C. M., & Winter, I. M. (2012). Speech enhancement inspired by auditory modelling. International Symposium on Hearing. • Hu, H., Mohammadiha, N., Taghia, J., Leijon, A., Lutman, M. E., Bleeck, S., & Wang, S. (2012). Sparsity Level in a Non-negative Matrix Factorization Based Speech Strategy in Cochlear Implants. EUSIPCO. • Li, G, Lutman, M. E., Wang, S., & Bleeck, S. (2012). Relationship between speech recognition in noise and sparseness. International Journal of Audiology, 51(2), 75–82. doi:10.3109/14992027.2011.625984 • Sang, J., Hu, H., Zheng, C., Li, G., Lutman, M. E., & Bleeck, S. (2012). Evaluation of a Sparse Coding Shrinkage Algorithm in Normal Hearing and Hearing Impaired Listeners. EUSIPCO (pp. 1–5). bleeck@gmail.com

More Related