US 11,771,372 B2
In-ear nonverbal audio events classification system and method
Jeremie Voix, Montreal (CA); Hami Montsarrat-Chanon, Montreal (CA); Rachel Bou Serhal, Montreal (CA); Patrick Cardinal, Kirkland (CA); and Philippe Chabot, Lasalle (CA)
Assigned to ECOLE DE TECHNOLOGIE SUPERIEURE, Montreal (CA)
Appl. No. 16/759,502
Filed by ECOLE DE TECHNOLOGIE SUPERIEURE, Montreal (CA)
PCT Filed Oct. 29, 2018, PCT No. PCT/CA2018/051369
§ 371(c)(1), (2) Date Apr. 27, 2020,
PCT Pub. No. WO2019/079909, PCT Pub. Date May 2, 2019.
Claims priority of provisional application 62/578,372, filed on Oct. 27, 2017.
Prior Publication US 2020/0312321 A1, Oct. 1, 2020
Int. Cl. A61B 5/00 (2006.01); G06N 20/00 (2019.01); G06F 3/16 (2006.01); G10L 15/22 (2006.01); G10L 25/51 (2013.01); G06F 18/213 (2023.01); G06F 18/21 (2023.01); G06F 18/241 (2023.01)
CPC A61B 5/6817 (2013.01) [G06F 3/16 (2013.01); G06F 18/213 (2023.01); G06F 18/217 (2023.01); G06F 18/241 (2023.01); G06N 20/00 (2019.01); G10L 15/22 (2013.01); G10L 25/51 (2013.01); G10L 2015/227 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for real time estimation of a noise exposure value in an ear of a user net of user-induced artefacts, the method comprising:
using a contactless in-ear microphone to capture an in-ear sound pressure present in an inner portion of an occluded ear canal of the user as an in-ear audio signal;
using an outer-ear microphone to capture an outer-ear sound pressure present at the outer entry of the occluded ear canal as an outer-ear audio signal wherein the outer-ear sound pressure is captured simultaneously to the captured in-ear sound pressure;
denoising the captured in-ear audio signal using the captured outer-ear audio signal;
sampling the in-ear audio signal;
extracting, in real time, user-induced non-verbal audio artefacts of each sample of the in-ear audio signal;
finding a match between each of the extracted user-induced non-verbal audio artefacts with one of a plurality of pre-recorded samples of user-induced artifacts:
subtracting the matching user-induced non-verbal artefact from the simultaneous in-ear audio signal:
estimating in-ear audio signal without user-induced artefacts using the remainder of the simultaneous in-ear audio signal:
calculating, in real-time, the noise-exposure value using the estimated signal.