Computational Audio Modelling for Robot-Assisted Assessment of Children’s Mental Wellbeing
cam.depositDate | 2022-10-24 | |
cam.issuedOnline | 2023-02-02 | |
cam.orpheus.counter | 9 | |
cam.orpheus.success | Tue Feb 14 10:21:06 GMT 2023 - This item is covered by RRS with an embargo. The item is now published and embargo has been lifted. | |
dc.contributor.author | Abbasi, NI | |
dc.contributor.author | Spitale, M | |
dc.contributor.author | Anderson, J | |
dc.contributor.author | Ford, T | |
dc.contributor.author | Jones, PB | |
dc.contributor.author | Gunes, H | |
dc.date.accessioned | 2022-10-25T23:30:32Z | |
dc.date.available | 2022-10-25T23:30:32Z | |
dc.date.issued | 2022 | |
dc.date.updated | 2022-10-24T17:05:10Z | |
dc.description.abstract | Robots endowed with the capability of assessing the mental wellbeing of children have a great potential to promote their mental health. However, very few works have explored the computational modelling of children’s mental wellbeing, which remains an open research challenge. This paper presents the first attempt to computationally assess children’s wellbeing during child-robot interactions via audio analysis. We collected a novel dataset of 26 children (8-13 y.o.) who interacted with a Nao robot to perform a verbal picture-based task. Data was collected by audio-video recording of the experiment session. The Short Mood and Feelings Questionnaire (SMFQ) was used to label the participants into two groups: (1)“higher wellbeing” (child SMFQ score <= SMFQ median), and (2) “lower wellbeing” (child SMFQ score > SMFQ median). We extracted audio features from these HRI interactions and trained and compared the performances of eight classical machine learning techniques across three cross-validation approaches: (1) 10 repetitions of 5-fold, (2) leave-one-child-out, and (3) leave-one-picture-out. We have also computed and analysed the sentiment of the audio transcriptions using the ROBERTa model. Our experimental results show that: (i) speech features are reliable for assessing children’s mental wellbeing, but they may not be sufficient on their own, and (ii) verbal information, specifically the sentiment that a picture elicited in children, may impact the children’s responses. | |
dc.description.sponsorship | This work was supported by the University of Cambridge’s OHMC Small Equipment Funding. N. I. Abbasi is supported by the W.D. Armstrong Trust PhD Studentship and the Cambridge Trusts. M. Spitale and H. Gunes are supported by the EPSRC project ARoEQ under grant ref. EP/R030782/1. All research at the Department of Psychiatry in the University of Cambridge is supported by the NIHR Cambridge Biomedical Research Centre (BRC-1215-20014, particularly T. Ford) and NIHR Applied Research Collaboration East of England (P. Jones, J. Anderson). | |
dc.identifier.doi | 10.17863/CAM.89924 | |
dc.identifier.eissn | 1611-3349 | |
dc.identifier.isbn | 9783031246692 | |
dc.identifier.issn | 0302-9743 | |
dc.identifier.uri | https://www.repository.cam.ac.uk/handle/1810/342506 | |
dc.language.iso | eng | |
dc.publisher | Springer Nature Switzerland | |
dc.publisher.department | Department of Computer Science and Technology | |
dc.publisher.url | http://dx.doi.org/10.1007/978-3-031-24670-8_3 | |
dc.rights | Attribution 4.0 International | |
dc.rights.uri | https://creativecommons.org/licenses/by/4.0/ | |
dc.subject | 46 Information and Computing Sciences | |
dc.subject | 4608 Human-Centred Computing | |
dc.subject | Mental Health | |
dc.subject | Pediatric | |
dc.subject | Brain Disorders | |
dc.subject | Behavioral and Social Science | |
dc.subject | Mental health | |
dc.subject | 3 Good Health and Well Being | |
dc.title | Computational Audio Modelling for Robot-Assisted Assessment of Children’s Mental Wellbeing | |
dc.type | Conference Object | |
dcterms.dateAccepted | 2022-10-14 | |
prism.publicationName | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | |
pubs.conference-finish-date | 2022-12-16 | |
pubs.conference-name | 14th International Conference on Social Robotics | |
pubs.conference-start-date | 2022-12-13 | |
pubs.funder-project-id | Engineering and Physical Sciences Research Council (EP/R030782/1) | |
pubs.funder-project-id | National Institute for Health and Care Research (NIHR200177) | |
pubs.funder-project-id | National Institute for Health and Care Research (IS-BRC-1215-20014) | |
pubs.licence-display-name | Apollo Repository Deposit Licence Agreement | |
pubs.licence-identifier | apollo-deposit-licence-2-1 | |
rioxxterms.version | AM | |
rioxxterms.versionofrecord | 10.1007/978-3-031-24670-8_3 |
Files
Original bundle
1 - 1 of 1
Loading...
- Name:
- ICSR_2022_Main_final.pdf
- Size:
- 7.08 MB
- Format:
- Adobe Portable Document Format
- Description:
- Accepted version
- Licence
- https://creativecommons.org/licenses/by/4.0/