Accepted Papers

Celebrate the researchers selected for presentation at the Child Speech AI Workshop. Explore the themes that emerged across the program and plan to meet the authors in Honolulu.

Program Reveal

Explore the accepted papers

Preview every oral and poster contribution, then check the schedule to see when each author presents.

  • Paper 1 of 6

    K-FUNCTION: JOINT PRONUNCIATION TRANSCRIPTION AND FEEDBACK FOR EVALUATING KIDS LANGUAGE FUNCTION

    Chenxu Guo1, Shuhe Li1, Jiachen Lian, Cheol Jun Cho, Elise Brenner, Francesca Pei, Maria Luisa Gorno-Tempini, Gopala Anumanchipalli

    Zhejiang University, UC Berkeley, UCSF

  • Paper 2 of 6

    Comparing Unsupervised and Supervised Semantic Speech Tokens: A Case Study of Child ASR

    Mohan Shi, Natarajan Balaji Shankar, Kaiyuan Zhang, Zilai Wang, Abeer Alwan

    University of California Los Angeles

  • Paper 3 of 6

    CAN LARGE AUDIO LANGUAGE MODELS UNDERSTAND CHILD STUTTERING SPEECH? SPEECH SUMMARIZATION, AND SOURCE SEPARATION

    Chibuzor Okocha, Maya Bakri and Christan Grant

    University of Florida, Lebanese American University

  • Paper 4 of 6

    Group-Aware Partial Model Merging for Children’s Automatic Speech recognition

    Thomas Rolland and Alberto Abad

    INESC-ID, Universidade de Lisboa

  • Paper 5 of 6

    TICL+: A CASE STUDY ON SPEECH IN-CONTEXT LEARNING FOR CHILDREN’S SPEECH RECOGNITION

    Haolong Zheng, Yekaterina Yegorova, Mark Hasegawa-Johnson

    University of Illinois at Urbana-Champaign

  • Paper 6 of 6

    BetterSpeak: An Atypical Speech to Typical Speech Platform for Dysarthric Speakers

    Seyed Reza Shahamiri

    The University of Auckland

Session logistics

Accepted papers session (oral)

All six accepted papers will be delivered as oral presentations in CC308, with the same slots reflected in the poster rotation.

  • Time: 16:50–18:20 (six back-to-back 15-minute slots).
  • Talk length: 10–12 minutes plus Q&A within each slot.
  • Ordering: papers scheduled by paper ID.
  • Mode: presentations are planned in person; remote delivery can be accommodated upon request.
  • One paper may be presented by a designated presenter who is not an author.

Thank you to everyone who contributed to this year’s call—we received our largest pool of submissions to date. After careful review, 6 papers were selected for presentation at ASRU 2025 in Honolulu. They represent advances across recognition accuracy, fairness, storytelling, and deployment in clinics and classrooms.

Authors will receive presentation logistics from the organizing committee. If you have questions about scheduling or accessibility needs, please reach out to jiachenlian@berkeley.edu.

Join the sessions

See when each paper is featured

Check the schedule for oral spotlights and poster rotations, and register to reserve your seat in Honolulu.