Inaugural International Workshop on Rethinking Children's Automatic Speech Recognition (ASR) for Education
In K–12 education, Automatic Speech Recognition (ASR) is increasingly being applied to support learning, engagement, and accessibility. However, developing ASR systems for children continues to poses monumental challenges, especially considering the stringent requirement resulting from various use cases in children’s education settings. Moreover, representative speech datasets from these populations are particularly scarce, while data collection is further complicated by ethical considerations such as privacy and informed consent.
Workshop Sponsors
Inaugural Workshop Chairs
Workshop Goals
This invitation only workshop provides a timely forum for interested stakeholders to come together to:
- Assess the state-of-the-art in children’s ASR
- Find a common ground in our collective understanding of the needs and challenges of children’s ASR in education
- Share lessons and best practices
- Standardize evaluation and benchmarking methodologies
- Charter new (or call-for-attention to important) research directions
- Build a coalition of shared minds to collectively address those challenges for the better development of children’s ASR for education settings
We invite participants from multidisciplinary communities, to join this discussion including educators, policymakers, ed-tech entrepreneurs, speech researchers, AI-ML experts, data scientists and linguists. Given your expertise in this area, we invite you to contribute to topics relevant to this workshop, including, but not limited to:
- Diversity and quality of existing public children’s speech data sets
- Ontology development for children’s speech datasets
- Equipment and systems for children’s speech data collection
- Evaluation methodologies for existing ASR systems on children’s speech
- Needs for multi-level ASR systems for children’s speech
- Educational applications spanning age groups, subject areas, language learning contexts, and delivery formats (e.g., in-person vs. virtual)
- Initiatives and requirements for making proprietary children’s speech data publicly accessible
- Current benchmark creation and dataset design efforts advancing ASR for children’s speech
- Privacy-preserving ASR and edge ASR
This workshop presents an exceptional opportunity for an interdisciplinary, cross-sector community to collectively shape the future of children’s ASR technologies in education. Together, we aim to develop novel AI solutions that can help improve learning outcomes for all students. Your participation and contribution will enrich these discussions and amplify our collaborative investment and efforts in building a better learning environment for our children.
Workshop Agenda
7:00 - 8:00 | Breakfast and Sign-in | |
8:00 - 8:15 | Welcome and Opening | UB Provost: Scott Weber UB SEAS Dean: Kemper Lewis |
8:15 - 9:00 | Scope of the Workshop | Jinjun Xiong (UB) |
9:00 - 10:00 | Challenges in Children’s ASR, Why Now? | Moderator: Jing Liu (UMD) Panelists: |
| ABSTRACT: Many assume that the challenges of ASR for children have already been solved. But is it really the case? Even with today’s advances in AI, significant hurdles remain. What are the persistent technical and practical obstacles? What innovative approaches might address them? And how does an educational focus on ASR technology influence these considerations? In this panel, experts with decades of experience in ASR will share insights into these complex issues, shed light on the roadblocks that still stand in the way, and offer pathways that can lead to breakthroughs moving forward. | ||
10:00 - 10:15 | Coffee Break | |
10:15 - 11:15 | Data Drives True AI, but Where do We Stand Today? | Moderator: Rohini Srihari (UB) Panelists: |
| ABSTRACT: The limited availability of open-source, large-scale speech datasets from children remains a major barrier to advancing AI-driven speech models for education. Privacy concerns make the collection and sharing of such data even more complex. This panel will showcase teams that are actively tackling these challenges by developing novel datasets of children’s speech. Leading scholars will present their approaches, share lessons learned and engage with the community to chart a path toward building a more robust data infrastructure, one that can accelerate progress in children’s ASR for educational applications. | ||
11:15 - 12:15 | Needs and Solutions: | Moderator: Brian Graham Panelists: |
| ABSTRACT: Many EdTech products rely on students’ voices as a primary input to power their core functionalities, whether it’s assessing mathematical reasoning, evaluating literacy development, or creating a seamless user experience. However, the requirements for children’s ASR vary widely depending on the subject domain, student subgroups, and the specific goals of each product. This panel brings together entrepreneurs from leading EdTech companies to share how they incorporate ASR into their tools, the unique needs and solutions they’ve developed, and their aspirations for the future of the field. The conversation is designed to spark greater alignment between academic research, industry needs, and funders’ investments, and ensure that progress in children’s ASR translates into meaningful impact in education. | ||
12:15 - 1:15 | LUNCH BREAK | |
1:15 - 2:15 | Children’s ASR in Action | Moderator: Marisha Speights (Northwestern University) Panelists: |
| ABSTRACT: Two flagship federal initiatives are leading the way in applying AI to education: the National AI Institutes program and the National Research and Development Centers on Using Generative Artificial Intelligence to Augment Teaching and Learning in Classrooms (U-GAIN). In this session, panelists representing these programs will share real-world educational use cases from their institutes and centers. They will discuss how children’s ASR is being integrated into their work, highlight successes and lessons learned, and reveal their “wish list” for future advancements in children’s ASR technology. | ||
2:15 - 3:15 | Interests from Funding Agencies and their Perspectives on Children’s ASR for Education | Moderator: Dana Suskind Panelists: |
| ABSTRACT: Funding plays a pivotal role in guiding the research priorities and efforts of the community. As the federal funding landscape evolves, support from private-sector foundations will become increasingly influential. In this discussion, program officers from both a federal funding agency (NSF) and private foundations will share their perspectives on the importance of children’s ASR in education. They will address its relevance within a broader, integrated solutions context, highlighting complementary technologies, and outline their “wish list” for future technology and solution development. | ||
3:15 - 3:30 | Coffee Break | |
3:30 - 4:30 | ASR and the Future | Moderator: Alison Hendricks (UB) Panelists: |
| ABSTRACT: Assessment is one of the most dynamic and promising frontiers for applying ASR in education. By enabling more individualized evaluation, ASR has the potential to transform traditional assessment systems by moving beyond rigid, end-of-year standardized tests toward approaches that are timely, adaptive and deeply informative. This panel brings together experts from leading assessment companies and research institutions to explore both the technical and practical dimensions of ASR-powered assessment. Together, they will share their insights on current innovations and offer their visions for the future of assessment in the age of AI. | ||
4:30 - 5:30 | Small Group Deep Dive | |
| ABSTRACT: After a full day of inspiring talks and lively conversations, attendees will have a chance to channel their energy and ideas into focused discussions. In small, topic-centered groups, participants will explore emerging key themes during the day, identifying areas worth pursuing beyond the event. Possible discussion topics include: (1) Standardizing evaluation and benchmarking for tracking progress in children’s ASR for education; (2) Sharing best practices and lessons learned across the community; (3) Coordinating collective data collection efforts to benefit all stakeholders, and (4) Building a lasting coalition to sustain momentum and drive continued impact. | ||
5:30 - 6:15 | Group Discussion for Next Steps | |
| ABSTRACT: We will bring everyone back together to share the key takeaways from each small-group discussion and collaboratively outline a roadmap for moving forward. To keep the momentum alive, we’ll conclude by inviting enthusiastic volunteer writers to lead the development of a white paper that captures our findings and recommendations for the broader community. | ||
6:15 - 6:30 | Travel to Dinner | |
6:30 - 8:30 | DINNER | |