The 1st Clarity Prediction Challenge: A machine learning challenge for hearing aid intelligibility prediction

Jon Barker; Michael Akeroyd; Trevor J. Cox; John F. Culling; Jennifer Firth; Simone Graetzer; Holly Griffiths; Lara Harris; Graham Naylor; Zuzanna Podwinska; Eszter Porter; Rhoddy Viveros Munoz

doi:10.21437/interspeech.2022-10821

The 1st Clarity Prediction Challenge: A machine learning challenge for hearing aid intelligibility prediction

Jon Barker
, Michael Akeroyd
, Trevor J. Cox
, John F. Culling
, Jennifer Firth
, Simone Graetzer
, Holly Griffiths
, Lara Harris
, Graham Naylor
, Zuzanna Podwinska
, Eszter Porter
, Rhoddy Viveros Munoz

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding

33 Citations (Scopus)

Abstract

This paper reports on the design and outcomes of the 1st Clarity Prediction Challenge (CPC1) for predicting the intelligibility of hearing aid processed signals heard by individuals with a hearing impairment. The challenge was designed to promote the development of new intelligibility measures suitable for use in developing hearing aid algorithms. Participants were supplied with listening test data compromising 7233 responses from 27 individuals. Data was split between training and test sets in a manner that fostered a machine learning approach and allowed both closed-set (known listeners) and open-set (unseen listener/unseen system) evaluation. The paper provides a description of the challenge design including the datasets, the hearing aid algorithms applied, the listeners and the perceptual tests. The challenge attracted submissions from 15 systems. The results are reviewed and the paper summarises, compares and contrasts approaches.

Original language	English
Title of host publication	Proceedings of Interspeech 2022
Pages	3508-3512
Number of pages	5
Volume	2022-September
DOIs	https://doi.org/10.21437/interspeech.2022-10821
Publication status	Published - 21 Sept 2022
Externally published	Yes
Event	Interspeech 2022 - Incheon, Korea, Democratic People's Republic of Duration: 18 Sept 2022 → 22 Sept 2022 https://www.interspeech2022.org/

Publication series

Name	Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
ISSN (Print)	2308-457X

Conference

Conference	Interspeech 2022
Country/Territory	Korea, Democratic People's Republic of
City	Incheon
Period	18/09/22 → 22/09/22
Internet address	https://www.interspeech2022.org/

Funding

Clarity is funded by UKRI (EP/S031448/1, EP/S031308/1, EP/S031324/1 and EP/S030298/1). We thank Amazon, the Hearing Industry Research Consortium and the Royal National Institute for the Deaf (RNID) for their support.

Funders	Funder number
UK Research and Innovation	EP/S031308/1, EP/S031324/1

Keywords

hearing aid
hearing loss
machine learning
speech intelligibility
speech-in-noise

ASJC Scopus subject areas

Software
Signal Processing
Language and Linguistics
Modelling and Simulation
Human-Computer Interaction

Access to Document

10.21437/interspeech.2022-10821

Cite this

Barker, J., Akeroyd, M., Cox, T. J., Culling, J. F., Firth, J., Graetzer, S., Griffiths, H., Harris, L., Naylor, G., Podwinska, Z., Porter, E., & Munoz, R. V. (2022). The 1st Clarity Prediction Challenge: A machine learning challenge for hearing aid intelligibility prediction. In Proceedings of Interspeech 2022 (Vol. 2022-September, pp. 3508-3512). (Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH). https://doi.org/10.21437/interspeech.2022-10821

The 1st Clarity Prediction Challenge: A machine learning challenge for hearing aid intelligibility prediction. / Barker, Jon; Akeroyd, Michael; Cox, Trevor J. et al.
Proceedings of Interspeech 2022. Vol. 2022-September 2022. p. 3508-3512 (Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH).

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding

Barker, J, Akeroyd, M, Cox, TJ, Culling, JF, Firth, J, Graetzer, S, Griffiths, H, Harris, L, Naylor, G, Podwinska, Z, Porter, E & Munoz, RV 2022, The 1st Clarity Prediction Challenge: A machine learning challenge for hearing aid intelligibility prediction. in Proceedings of Interspeech 2022. vol. 2022-September, Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH, pp. 3508-3512, Interspeech 2022, Incheon, Korea, Democratic People's Republic of, 18/09/22. https://doi.org/10.21437/interspeech.2022-10821

Barker J, Akeroyd M, Cox TJ, Culling JF, Firth J, Graetzer S et al. The 1st Clarity Prediction Challenge: A machine learning challenge for hearing aid intelligibility prediction. In Proceedings of Interspeech 2022. Vol. 2022-September. 2022. p. 3508-3512. (Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH). doi: 10.21437/interspeech.2022-10821

@inproceedings{c851152f970549e1ab766964931b1675,

title = "The 1st Clarity Prediction Challenge: A machine learning challenge for hearing aid intelligibility prediction",

abstract = "This paper reports on the design and outcomes of the 1st Clarity Prediction Challenge (CPC1) for predicting the intelligibility of hearing aid processed signals heard by individuals with a hearing impairment. The challenge was designed to promote the development of new intelligibility measures suitable for use in developing hearing aid algorithms. Participants were supplied with listening test data compromising 7233 responses from 27 individuals. Data was split between training and test sets in a manner that fostered a machine learning approach and allowed both closed-set (known listeners) and open-set (unseen listener/unseen system) evaluation. The paper provides a description of the challenge design including the datasets, the hearing aid algorithms applied, the listeners and the perceptual tests. The challenge attracted submissions from 15 systems. The results are reviewed and the paper summarises, compares and contrasts approaches.",

keywords = "hearing aid, hearing loss, machine learning, speech intelligibility, speech-in-noise",

author = "Jon Barker and Michael Akeroyd and Cox, \{Trevor J.\} and Culling, \{John F.\} and Jennifer Firth and Simone Graetzer and Holly Griffiths and Lara Harris and Graham Naylor and Zuzanna Podwinska and Eszter Porter and Munoz, \{Rhoddy Viveros\}",

year = "2022",

month = sep,

day = "21",

doi = "10.21437/interspeech.2022-10821",

language = "English",

volume = "2022-September",

series = "Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH",

pages = "3508--3512",

booktitle = "Proceedings of Interspeech 2022",

note = "Interspeech 2022 ; Conference date: 18-09-2022 Through 22-09-2022",

url = "https://www.interspeech2022.org/",

}

TY - GEN

T1 - The 1st Clarity Prediction Challenge: A machine learning challenge for hearing aid intelligibility prediction

AU - Barker, Jon

AU - Akeroyd, Michael

AU - Cox, Trevor J.

AU - Culling, John F.

AU - Firth, Jennifer

AU - Graetzer, Simone

AU - Griffiths, Holly

AU - Harris, Lara

AU - Naylor, Graham

AU - Podwinska, Zuzanna

AU - Porter, Eszter

AU - Munoz, Rhoddy Viveros

PY - 2022/9/21

Y1 - 2022/9/21

N2 - This paper reports on the design and outcomes of the 1st Clarity Prediction Challenge (CPC1) for predicting the intelligibility of hearing aid processed signals heard by individuals with a hearing impairment. The challenge was designed to promote the development of new intelligibility measures suitable for use in developing hearing aid algorithms. Participants were supplied with listening test data compromising 7233 responses from 27 individuals. Data was split between training and test sets in a manner that fostered a machine learning approach and allowed both closed-set (known listeners) and open-set (unseen listener/unseen system) evaluation. The paper provides a description of the challenge design including the datasets, the hearing aid algorithms applied, the listeners and the perceptual tests. The challenge attracted submissions from 15 systems. The results are reviewed and the paper summarises, compares and contrasts approaches.

AB - This paper reports on the design and outcomes of the 1st Clarity Prediction Challenge (CPC1) for predicting the intelligibility of hearing aid processed signals heard by individuals with a hearing impairment. The challenge was designed to promote the development of new intelligibility measures suitable for use in developing hearing aid algorithms. Participants were supplied with listening test data compromising 7233 responses from 27 individuals. Data was split between training and test sets in a manner that fostered a machine learning approach and allowed both closed-set (known listeners) and open-set (unseen listener/unseen system) evaluation. The paper provides a description of the challenge design including the datasets, the hearing aid algorithms applied, the listeners and the perceptual tests. The challenge attracted submissions from 15 systems. The results are reviewed and the paper summarises, compares and contrasts approaches.

KW - hearing aid

KW - hearing loss

KW - machine learning

KW - speech intelligibility

KW - speech-in-noise

UR - https://www.scopus.com/pages/publications/85128551580

U2 - 10.21437/interspeech.2022-10821

DO - 10.21437/interspeech.2022-10821

M3 - Conference proceeding

VL - 2022-September

T3 - Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH

SP - 3508

EP - 3512

BT - Proceedings of Interspeech 2022

T2 - Interspeech 2022

Y2 - 18 September 2022 through 22 September 2022

ER -

The 1st Clarity Prediction Challenge: A machine learning challenge for hearing aid intelligibility prediction

Abstract

Publication series

Conference

Funding

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this