Automated L2 Speaking Assessment (AL2SA) International Workshop 2026

THURSDAY, MARCH 5TH

Session n°1: Assessing speaking skills

9:20-9:50 Linda Terrier & Lionel Fontan 
Univ. Toulouse Jean-Jaurès, France
Archean Labs, France
What Are We Actually Measuring? Reflections on the Construct of Intelligibility in Automated L2 Speech Assessment

9:50-10:20 Nivja de Jong
Leiden Univ., the Netherlands
What is speaking proficiency and how to develop high-quality, practical, and ethical automated assessments for its measurement

10:20-11:00 Discussion

Session n°2: Listening Disfluency

11:30-12:10 Nobuaki Minematsu
Univ. of Tokyo, Japan
Measuring, Analyzing, and Predicting Listening Disfluency of Learners and Raters: Using Speech and AI Technologies for Automated Assessment

12:10-12:30 Noriko Nakanishi (online)
Kobe Gakuin Univ., Japan
The Shadowing Exchange Community: Enhancing Accent Perception, Intelligible Speech, and Empathetic Feedback

12:30-13:00 Discussion

Session n°3: Teachers' Open Session

14:30-15:30 Beata Walesiak
Unpolish.pl, Poland
Apps for L2 pronunciation training

15:30-16:00 Sylvain Coulange & Pinxun Huang
Univ. Grenoble Alpes, France
Univ. Lorraine, France
Designing a Speaking Assessment Module for the SELF Language Placement Test

16:00-16:30 Discussion

FRIDAY, MARCH 6TH

Session n°4: Intelligibility

9:00-9:30 Dan Frost
Univ. Grenoble Alpes, France
Addressing language-specific needs: what makes learner speech intelligible and how can we assess it?

9:30-10:00 Kevin Hirschi
Univ. of Texas San Antonio, USA
Towards automated measurement and feedback of L2 intelligibility: Challenges and a pedagogically informed roadmap

10:00-10:30 Joan Carles Mora
Univ. de Barcelona, Spain
Measures of acoustic and perceptual contrastiveness and nativelikeness in assessing segmental pronunciation development

10:30-11:00 Discussion

Session n°5: Interactions

11:30-12:00 Serge Bibauw
Univ. Catholique de Louvain, Belgium
Conversational AI for spoken L2 development: meta-analysis of effectiveness studies and insights for assessment

12:00-12:30 Tsuneo Kato
Doshisha Univ., Japan
Effect of Prompt Corrective Feedback and Analysis of Error Patterns in Learning Syntactic Form with Trialogue-based CALL System

12:30-13:00 Mayuko Aiba
Univ. of Tokyo, Japan
(Title to be added)

13:00-13:30 Discussion

Session n°6: LLM-Based Assessment

15:00-15:30 Nicolas Ballier
Univ. Paris Cité, France
What's in the L2 speech signal? Calibrating Whisper probability scores with phonetic posteriorgrams

15:30-16:00 Stefano Bannò
Univ. of Cambridge, UK
Natural Language-based Assessment of L2 Oral Proficiency using LLMs

16:00-16:30 Luis Da Costa & Laura Rupp
Vrije Univ. Amsterdam, the Netherlands
Perspectives for Large Scale Speaking Assessment within the Mooc “English Pronunciation in a Global World”

16:30-17:00 Discussion