John Mendonça is a Junior Researcher of the HLT (Human Language Technologies Laboratory) and a PhD Student of the Doctoral Programme in Electrical and Computer Engineering of IST. He obtained his MSc degree in 2020 with the thesis "Automatic Detection of Profile Features". In it, he applied speech processing techniques, more specifically in speaker recognition and paralinguistics, for the collection of crowdsourced and health related datasets. More recently he joined a PhD Programme under the MAIA project, a joint CMU/INESC-ID/IT/Unbabel collaboration for Intelligent AI Agents in chatbots. His PhD work is closely related to conversational quality estimation.
- - Email: John.mendonca(at)inesc-id.pt
- - Role: Junior Researcher | PhD Student
- - Human Language Technologies Lab, INESC-ID | Instituto Superior Técnico
- - Rua Alves Redol, 9, 1000-029 Lisboa, Portugal
- - inesc-id.pt | tecnico.ulisboa.pt
In 2020, he joined the HLT team for his Master Thesis Research. During this time, he led the HLT Team's contribution to the ComParE 2020 Breathing sub-challenge, pertaining the estimation of breathing signals. In 2021, his research focus shifted to Natural Language Processing (NLP), with special emphasis in conversational agent evaluation metrics.
- John Mendonça, Alon Lavie, Isabel Trancoso, QualityAdapt: an Automatic Dialogue Quality Estimation Framework, In 23rd Annual Meeting of the Special Interest Group on Discourse and Dialogue, Association for Computational Linguistics, pages 83-90, Edinburgh, September 2022
- John Mendonça, Rui Correia, Mariana Lourenço, João Freitas, Isabel Trancoso, Towards Speaker Verification for Crowdsourced Speech Collections, In Proceedings of the Language Resources and Evaluation Conference, European Language Resources Association, pages 5929--5937, Marseille, France, June 2022
- John Mendonça, Francisco Teixeira, Isabel Trancoso, Alberto Abad, Analyzing Breath Signals for the Interspeech 2020 ComParE Challenge, In Interspeech 2020, October 2020