EVALUATION OF A COMPLEMENTARY HEARING AID FOR SPATIAL SOUND - PowerPoint Presentation

1. EVALUATION OF A COMPLEMENTARY HEARING AID FOR SPATIAL SOUNDSEGREGATIONLuca Giuliani1, Luca Brayda1, Sara Sansalone2, Stefania Repetto2 and Michele Ricchetti2 Fondazione Istituto Italiano di Tecnologia, Genoa, Italy2 Linear s.r.l, Genoa, Italy1

2. OutlineThe cocktail party problemBeamformingState of the artThe Glassense systemExperimental setup and protocolData analysis and resultsConclusion and future works2

3. The cocktail party problemFirstly defined by Colin Cherry in 1953, consists in the trouble of perceiving speech in noisy social contests.Colin Cherry3

4. The cocktail party problemApproximately one-third of people over 65 year are affected by disabling hearing loss.People suffering from hearing impairment have bigger troubles in separating acoustical sources.Limited hearing aids performance in noise.4

5. BeamformingAlso called spatial filtering, is a signal processing technique used in sensor arrays to improve the Signal to Noise Ratio (SNR) of a target source.5

6. State of the artModern behind-the-ear earing aids are equipped with more then one microphone to provide directionality.The improvement deeply depends on the distance between the microphones and on their number.6

7. State of the artSeveral solutions has been proposed in the last years, but none of them made a real breakthrough in the market yet.Hearing-aid necklace by Widrow (2001)7High directivityComplementary to hearing aidNo need for calibration on hearing lossCannot use head as spatial selectorNot available on the market anymore

8. State of the artSeveral solutions has been proposed in the last years, but none of them made a real breakthrough in the market yet.8High directivityUse head as spatial selectorSubstitutes hearing aidNeed for calibration on personal hearing lossHearing glasses by Varibel company (available on the market since 2008)

9. The GlassenseSpectacles with two arrays of digital microphones embedded in the temples.9Signal sent to hearing aids wireless

10. The GlassenseEach temple embeds a 10 cm array of four digital microphones.10

11. The GlassenseBeam power pattern at various frequencies. The sounds coming from the sides and the rear are attenuated.11

12. The Glassense12The white noise gain (WNG) is defined as the ability to suppress spatially uncorrelated noise

13. Is it possible to overcome spatial filtering limitations of traditional hearing aids to improve speech comprehension capabilities for hearing impaired people in noisy environments?13

14. Subjects8 subjects6 hearing impaired2 healthy hearingAge 59-78 years14

15. Experimental SetupAnechoic roomFour speakers around the subject, 1m away at 0° (directly ahead), 180° (back) and ±90°All subjects wore a couple of in-ear hearing aids15

16. Experimental SetupTarget signal: Lists of 10 meaningful bisyllabic unrelated words (200 in total)Reproduced by the frontal speaker16

17. Experimental SetupCompeting noise:Four channel registration of a cocktail-like typical acoustic backgroundReproduced by all the surrounding speakers (included the frontal one)17

18. Experimental ProtocolTrial = listen and repeat task on target words with competing noise Performance = % of correctly repeated wordsTarget signal calibrated on the subject to reach 100% performance without noise18

19. Experimental VariablesIndependent variable: SNR LevelTarget speech volume fixedCompeting noise volume changed in each trial, i.e. different Signal to Noise Ratios (SNRs) Dependent variable: Performance19

20. Experimental ConditionsHearing aids conditionsounds perceived directly by hearing aids microphones.20

21. Experimental ConditionsUnfiltered conditionGlassense connected to hearing aidssounds perceived by Glassense microphones, without spatial filtering21

22. Experimental Conditions22Beamforming conditionGlassense connected to hearing aidssounds perceived by Glassense microphones, with spatial filtering

23. Data AnalysisSpeech Reception Threshold (SRT)SNR at which the subject correctly understand 50% of target wordsObtained interpolating subjects performances23

24. Data AnalysisFriedman Test :Dependent variable = SRTIndependent variable = Listening conditionPost Hoc:Wilcoxon-Mann-WhitneyBonferroni correction24

25. Data analysis and resultsMean SRT comparisonimprovement of 3.3 dB in beamforming condition with respect to unfiltered improvement of 2.4 dB in beamforming condition with respect to hearing aids25

26. Data analysis and resultsMean SRT comparisonimprovement of 3.3 dB in beamforming condition with respect to unfiltered improvement of 2.4 dB in beamforming condition with respect to hearing aids26

27. ConclusionThe use of Glassense can improve the performance of traditional hearing aids in noise.27

28. ConclusionNot easy to compare Glassense SRT improvement to other devices (like Varibel Hearing Glasses1) because of differences in experimental protocol.[1] - L. H. M. Mens, “Speech understanding in noise with an eyeglass hearing aid: asymmetric fitting and the head shadow benefit of anterior microphones.”, Int. J. Audiol., vol. 50, no. 1, pp. 27–33, 2011.28

29. Conclusion29No need for calibration on hearing loss, since hearing aid is already calibrated.Easy to swap between traditional hearing aid and Glassense.

30. Future worksExtension of experimental populationMeasurement of Glassense contribution with more hearing aid modelsComparison with other microphone array devices30

31. Thank you!Francesco Diotalevi Senior TechnicianLuca BraydaTeam LeaderLinear s.r.l.Michele RicchettiProject CoordinatorSara SansaloneStefania RepettoPetra BianchiClaudio LoriniDavide DellepianeElio MassaCo-funding:31

32. Questions?32

33. EVALUATION OF A COMPLEMENTARY HEARING AID FOR SPATIAL SOUNDSEGREGATIONluca.giuliani@iit.it/IITGlassense@IITGlassense33DEMO AVAILABLE!

34. Openings @ IITPost-doc position Antropometric modeling for personalized sensory aidsPhD position Binaural filtering techniques for augmented perception and understanding of sounds 34Email to: luca.brayda@iit.it