The Social and Technical Values of a Grassroots Chinese Stuttered Speech Dataset
December 22, 2025
Season’s greetings from all of us at AImpower.org!
December was a busy and energizing month for our team. We published an ACM TechBriefs article examining the ethical impacts of automatic speech recognition (ASR); wrote two blog posts summarizing our research on stuttered speech data collection and governance; prepared for two upcoming academic workshops (InterSpeech, CHI) focused on inclusive speech AI; and wrapped up our month-long beta test for LibOrate — with overwhelmingly positive feedback from the stuttering community ❤️.
As we reflect on 2025, we see it as a year of execution and amplification. While doubling down our technical endeavor to make AI and videoconferencing more inclusive and empowering for people who stutter, we also strategically expanded our impact through capacity-building and partnerships. Working closely with the global stuttering community across the United States, Canada, the United Kingdom, China, and Japan, we shipped technical artifacts that catalyze change within the AI ecosystem and advocated for speech inclusivity across academic, technical, media, and philanthropy spaces — bringing more people into the movement everyday.
Still a young and small organization — with only two staff members and a dozen of contributors and volunteers — we are incredibly proud of what we achieved in 2025 with your support. Amid heightened external pressure and uncertainty, we are deeply grateful to our institutional funders, including the National Science Foundation, the Patrick J. McGovern Foundation, and Borealis Philanthropy. Above all, we want to thank our community, collaborators, and friends for supporting us with your data, ideas, and connections. We look forward to working even more closely with you — and hopefully meeting some of you in person — in 2026.
Here’s to joyful holidays and a 2026 filled with hopes and solidarity!
Warmly,
Shaomei @ AImpower.org
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.
2025 In Review: By The Numbers
Here are a few numbers that reflect our technical outputs, community mobilization, and public advocacy efforts in 2025.
Highlights From AImpower.org in 2025
1. Building a cross-sector, cross-country coalition to amplify marginalized voices in tech
We do not change the world alone. 2025 taught us the importance of solidarity, partnership, and collaboration. We successfully hosted the very first "Speech AI for All" workshop at CHI 2025, with a great turn-out of cross-sector participants from industry, academic, and community organizations. We had also formed a working group and would continue organizing other events, activities, and collaborations with this amazing group of people who shared our vision and goals to build "speech AI for all"!
Speech AI for All workshop @ CHI 2025
Collaboration with Patrick J McGovern Foundation
Partnership with Speech Accessibility Project
2. Working directly with hundreds of people who stutter to advocate for and build inclusive communication environments for all
We are proud of being part of, and working closely with, the stuttering community. We continued our commitment to stuttering awareness and advocacy in 2024, and had lots of fun and inspiring conversations with our old or new friends who stutter at National Stuttering Association's annual conference at Denver, as well as the Friends conference at Charlotte.
Stuttering Advocacy
Community events
3. Producing high quality scientific research on inclusive speech AI
We published six papers at top conferences such as CHI, FAccT, AIES, InterSpeech, and ACL in 2025, all thanks to the great team of our staff, volunteers, collaborators, and partners. One of our papers won the best paper award at the AAAI/ACM Conference on Artificial Intelligence, Ethics, and Society (AIES '25)!
Here are our academic publications in 2025:
-
Speech AI for All: Promoting Accessibility, Fairness, Inclusivity, and Equity. Shaomei Wu, Kimi V. Wenzel, Jingjin Li, Qisheng Li, Alisha Pradhan, Raja Kushalnagar, Colin Lea, Allison Koenecke, Christian Vogler, Mark Hasegawa-Johnson, Norman Makoto Su, and Nan Bernstein Ratner. CHI Workshop Proposal. In Extended Abstracts of the CHI Conference on Human Factors in Computing Systems (CHI EA ’25). 2025. [pdf] [workshop website]
-
Our Collective Voices: The Social and Technical Values of a Grassroots Chinese Stuttered Speech Dataset. Jingjin Li, Qisheng Li, Rong Gong, Lezhi Wang, and Shaomei Wu. In Proceedings of the ACM Conference on Fairness, Accountability, and Transparency (FAccT '25). Pages 2768 - 2783. doi:10.1145/3715275.3732179. 2025. [preprint]
-
Lost in Translation: Benchmarking Commercial Machine Translation Models for Dyslexic-Style Text. Gregory Price and Shaomei Wu. In Findings of the Association for Computational Linguistics: ACL 2025, pages 13771–13782, Vienna, Austria. Association for Computational Linguistics. [preprint]
-
J-j-j-just Stutter: Benchmarking Whisper’s Performance Disparities on Different Stuttering Patterns. Charan Sridhar, Shaomei Wu. Proc. InterSpeech Conference 2025, 3753-3757, doi: 10.21437/Interspeech.2025-2700 [preprint, presentation]
-
Govern With, Not For: Understanding the Stuttering Community’s Preferences and Goals for Speech AI Data Governance in the US and China. Jingjin Li, Peiyao Liu, Rebecca Lietz, Ningjing Tang, Norman Makoto Su, and Shaomei Wu. In Proc. of the Eighth AAAI/ACM Conference on Artificial Intelligence, Ethics, and Society (AIES '25). 2025. Best Paper Award [preprint)
-
ACM TechBrief: Automated Speech Recognition. Allison Koenecke, Niranjan Sivakumar, Jingjin Li, Shaomei Wu. ACM TechBriefs, Winter 2025, Issue 15. December 11, 2025.
AIES 2025 Best Paper Award
4. Driving policy conversations around Automatic Speech Recognition technology
To drive real changes in speech technology for people with speech diversities, we started working with leading academic researchers in this field to publish scientific statistic and briefings for policy makers, and learned a great deal during this process.
ACM Tech Brief on Automatic Speech Recognition Technology
5. Delivering multiple tangible technical products to support and benefit people who stutter in technology-mediated communications
We beta launched our Zoom companion app, LibOrate, in late 2025, with a warm reception from community beta testers. Our ASR related outputs - the stuttered speech dataset and the ASR model fine-tuned with stuttered speech - continued more and more attention and had been downloaded hundreds of times in 2025!
LibOrate - Zoom companion app for people who stutter
Stuttering-friendly Open ASR model
Mandarin Stuttered Speech Dataset
Looking Ahead to 2026
We have exciting plans for 2026 ✨. We will continue building bridges by organizing academic workshops and presenting at major stuttering community conferences in the U.S. and abroad. We are launching a new research collaboration with Chris Constantino and the University of Tsukuba in Japan, publicly releasing LibOrate on the Zoom Marketplace, and publishing our English Stuttered Speech Dataset alongside an upgraded inclusive ASR model trained on this data.
Together, we’re building a global movement to transform speech technology and ensure every voice is heard. Join us!
