Alkis Koudounas

Ph.D. in Ingegneria Informatica E Dei Sistemi , 38th cycle (2022-2025)

Ph.D. obtained in 2026

Dissertation:

Toward Robust, Responsible and Trustworthy Speech Foundation Models

Tutors:

Elena Maria Baralis Eliana Pastor

Research presentation:

Video presentation Poster

Profile

Research topic

Toward Robust, Responsible and Trustworthy Speech Foundation Models

Research interests

Data science, Computer vision and AI

Biography

Alkis is a third-year Ph.D student at the Polytechnic University of Turin, Italy, and an applied research intern at Amazon AGI. His research focuses on speech, audio, and multimodal understanding, as well as robust, responsible, and trustworthy AI. His interest also lies in the development of resources for under-represented languages, and he is currently focusing on creating tools to help diagnose voice-related pathologies and improve the quality of life for those affected by them. He served as the Italian Language Ambassador for the AYA Cohere4AI project. His works have been published in top-tier speech and NLP conferences and journals, including ICASSP, ACL, IEEE/ACM TASLP, COLING, EACL, and Interspeech, where he also won the Best Student Paper Award (2024).

Personal website: https://koudounasalkis.github.io/

Awards and Honors

  • Participation in the Symposium on Artificial Intelligence (https://synapsesymposium.ai/) and poster presentation of the PhD research. (2023)
  • Italian and Greek Language Ambassador for the AYA Cohere4AI project (01/09/2023 - 31/12/2024) (2024)
  • IEEE ICASSP 2024 Travel Grant Recipient (2024)
  • ACM KDD 2024 Travel Grant Recipient (2024)
  • ISCA Interspeech 2024 Travel Grant Recipient (2024)
  • Best Student Paper Award at the ISCA Interspeech 2024 for the paper "A Contrastive Learning Approach to Mitigate Bias in Speech Models" (https://youtu.be/Kn4zScqw2ro?si=Kc2u56uO0m07CDMZ) (2024)
  • Online Chair at ECML-PKDD 2023 (2023)
  • Applied Scientist Intern at Amazon AGI - Speech Recognition Team (2024)
  • Joint Project (04/2025 - ongoing) with Amazon AGI on "SpeechLLMs Preference Modeling", Lead Author and Principal Author of the Proposal. (2025)
  • Participation in the Mediterranean Machine Learning school (https://www.m2lschool.org/) from 11th to 16th September 2022. (2022)
  • Participation in the Speech and Natural Language Processing Winter school in the Alps (ALPS, https://lig-alps.imag.fr/) from 16th to 20th January 2023, and poster presentation of the PhD research. (2023)
  • Participation in the Generative Modeling Summer School (GEMSS, https://gemss.ai/2023/) from 26th to 30th June 2023, and poster presentation of the PhD research. (2023)
  • SmartTalk at SmartData@PoliTO center on PhD research, 11th March 2024 (https://smartdata.polito.it/exploring-subgroup-performance-in-end-to-end-speech-models/). (2024)
  • PhD's Pitch @ IEEE Polito Student Branch Title: “Subgroup Disparities in Speech Models: Detection and Mitigation” Short Abstract: The research addresses performance disparities in speech models across subgroups through in-processing (divergence-aware regularization, targeted augmentation, contrastive learning) and post-processing (targeted data acquisition) techniques, improving robustness for underrepresented demographics. (2025)
  • Joint Project (03/2022 - 03/2024) with Amazon Alexa AI on "Explaining Model Bias and Behavior for End-to-End SLU Models", Lead Author. Participation in the Amazon-PoliTo Workshops on March 23, 2023, and March 01, 2024, and presentation of project results. (2024)
  • Joint Project (04/2024 - 03/2025) with Amazon Alexa AI on "LLM Prompting in Multimodal Time-Evolving Scenarios", Author. Participation in the Amazon-PoliTo Workshop on March 07, 2025, and presentation of project results and ongoing research. (2025)
  • Participation at the A&T Fair from 14th February to 16th February 2024, volunteering at the stand DAUIN-PoliTO, and poster presentation of the PhD research. (2024)
  • Invited Talk at Amazon AGI - Speech Recognition and Understanding Team, on Research Highlights from PhD, 28th March 2025. (2025)
  • Teaching activities (22-23, 23-24, and 24-25) on: 1) Data Science e Tecnologie Basi di Dati, Master, 8 CFU: EL+TU 24h (22-23), EL 18h (23-24), TU 39h (24-25) 2) Data Science Lab: Process and Methods, Master, 8 CFU: EL+TU 36h (22-23), EL+TU 42h (23-24), TU 21h (24-25) Latest CPD: 3.60/99.23% (DSTBD) and 3.32/94.76% (DSL) (2025)
  • Invited Talk at the School of AI Algiers about "Subgroup Performance in End-to-End Speech Models" on the 10th November 2023. (2023)
  • Co-organizer of the Speech Pathology Analysis and DEtection (SPADE) Workshop, co-located with the IEEE ICASSP 2025 Conference (https://spadeworkshop.github.io/). (2025)
View moreView less

Teaching

Teachings

Master of Science

MostraNascondi A.A. passati

Research

Research groups

Publications

Works published during the Ph.D. View all publications in Porto@Iris

More publicationsLess publications