Tech4ALL Digest, Mar 11
March 18, 2025
A New Partnership to Empower Stuttering Voices in Speech AI
July 1, 2025AImpower.org just celebrated its third birthday – we are a toddler now! š„³
As we enter our fourth year of operations, we want to first reflect on the tremendous strides AImpower.org has made in 2024 in our mission to co-design empowering technologies for, with, and by marginalized communities. The year of 2024 has been a period of exponential growth in organizational capacity, research, product development, and advocacy. Here are some highlights from an extraordinary year.
Growing Our Organization
Fundraising
AImpower.org saw a 20x growth in fundraising in 2024, securing critical capital to support our research, product development, and advocacy initiatives to address technological challenges for people with verbal diversities. We thank our major funders and supporters below:
- National Science Foundation (NSF): Partnering with Professor Norman Makoto Su from University of California, Santa Cruz, we received a research award from NSF’s prestigious Responsible Design, Development, and Deployment of Technologies (ReDDDoT) program to fund our work on āDestigmatizing Disfluencies in Speech AI with Grassroots Stuttering Communitiesā. We have been leveraging the support from this award to investigate methodological biases in speech science and technology research and further engage the stuttering community in designing and developing disfluency-friendly speech AI technologies.
- Patrick J. McGovern Foundation (PJMF): We are honored to receive a generous grant from PJMF to support our work on developing fair and accessible automatic speech recognition models with the stuttering community. We look forwards to delivering a community-created stuttered speech dataset in 2025 that unlocks the much needed technological development in this area.
- San Francisco Foundation: We are grateful for a general development grant from the San Francisco Foundation that funds the additional operational infrastructure that allowed us to grew and expanded our programs;
- Borealis Philanthropy: We are extremely proud to be one of the new Disability Inclusion Fund grantees in 2024-2025 and deeply appreciate the support and recognition from Borealis, Ford Foundation, and MacArthur Foundation for disability-affirmative technologies we are building.
- Personal donations & corporate matching: As always, I want to personally thank the tens of individuals who have donate their time and/or money to our organization. Thank you for believing in us! AImpower.org would not exist today without your support. ā¤ļø
Personnel
We also had some exciting development within our team in 2024!
We finally had two full-time employees – a 100% growth from 2023! Jingjin Li, our newest staff member, has been a long time volunteer since 2023, leading our participatory research and design on inclusive videoconferencing technology. We are very excited to have her formally join AImpower.org and have an even broader impact on our projects and operations.
We also welcomed a new board member – Kaiton Williams! Kaiton oversees investments at Impact America Fund and has offered invaluable support to AImpower.org since our inception. Benefiting from Kaiton’s experience in impact technology and fundraising for the past three years, we are thrilled to formally have him join in our board, and look forwards to working even more closely with Kaiton on our product and organizational strategy in 2025.
AImpower.org can not exist without the labor of love from our volunteers. As a small organization with big goals, our volunteers and student interns enable us to make significant progress in research and technology development despite having only two staff members. We have over 30 contributing volunteers during 2024, mostly students and stuttering community members attracted by our mission. As we grew and expanded in 2024, we were particularly thankful for the dedicated support provided by Rongcheng Zhang, our new volunteer head of operations and marketing. A uncapped hero, Rongcheng has been picking up a wide range of tasks, including but not limited to book keeping, website maintenance, newsletter design and distribution, Google ads distribution, social media marketing..
Partnerships
AImpower.org continued to expand our network and influence, plugging ourselves into an ever growing ecosystem of public interest technology developers and grassroots, community movement builders. Besides our funders and collaborators, we have expanded and deepened our partnership with universities (e.g. Cornell University, University Ottawa, University of California, Santa Cruz, Boston University, Stanford University), sponsoring internship programs, hackathons, and guest lectures to engage and equip the next generation of technology designers and builders with a justice-oriented mindset. We also fostered our ties with other community organizations, such as SPACE and StammerTalk, building a new coalition at the intersection of stuttering and technology within the stuttering community. We also joined force with prominent professional and support organizations – such as American Speech-Language-Hearing Association (ASHA) – in advocating for the inclusion of diverse voices in science and in our society.
Researching Technological Tension with Stuttering
As a key pillar of AImpower.org work, our research foregrounds the lived experiences of the impacted community to amplify their agency and creativity in identifying and addressing their technological challenges and needs. Since day 1, we have been actively sharing our learnings in academic publications, reports, or our blog posts to maximize transparency and catalyze the movement for systematic change.
Thanks to the collective efforts from our volunteers, staff, and community partners, our research output grew 5x compared in 2023, with five publications in top-tier academic conferences, solidifying our thought leadership in speech AI and videoconferencing accessibility:
- Shaomei Wu, Jingjin Li, and Gilly Leshed. Finding My Voice over Zoom: An Autoethnography of Videoconferencing Experience for a Person Who Stutters. CHI ’24 full paper.
- Qisheng Li and Shaomei Wu. Towards Fair and Inclusive Speech Recognition for Stuttering: Community-led Chinese Stuttered Speech Dataset Creation and Benchmarking. CHI ā24 Late Breaking Work
- Jingjin Li, Shaomei Wu, and Gilly Leshed. Re-envisioning Remote Meetings: Co-designing Inclusive and Empowering Videoconferencing with People Who Stutter. ACM Designing Interactive Systems Conference (DIS ’24). Honorable Mention š Ā
- Rong Gong, Hongfei Xue, Lezhi Wang, Xin Xu, Qisheng Li, Lei Xie, Hui Bu, Shaomei Wu, Jiaming Zhou, Yong Qin, Binbin Zhang, Jun Du, Jia Bin, Ming Li. AS-70: A Mandarin stuttered speech dataset for automatic speech recognition and stuttering event detection. InterSpeech 2024.
- Qisheng Li and Shaomei Wu. “I Want to Publicize My Stutter”: Community-led Collection and Curation of Chinese Stuttered Speech Data. ACM Conference on Computer-Supported Cooperative Work and Social Computing (CSCW ’24).
Developing Community-Centric Technical Products
Beyond delivering rigorous scientific findings, AImpower.org also expects and cares deeply about creating tangible, sustainable interventions that bring positive impact on the lives of the communities we work with. To meet this goal, we have developed and released two technical products for inclusive videoconferencing and fair speech AI:
- LibOrate: Co-designed with the stuttering community, LibOrate is a videoconferencing companion app that aims to enhance the socio-emotional experience of meetings and facilitates a supportive meeting culture for all meeting participants. We have developed a working prototype of LibOrate in 2024 and expect to beta launch it in H1 2025. Check out its official sites for news and updates. The app is also open sourced, with majority of the code contributions came from the stuttering community.
- StammerTalk dataset: the first and largest Mandarin stuttered speech dataset created by the StammerTalk community in partnership with AImpower.org, featuring 72 speakers who stutter and nearly 50 hours of stuttered speech data with both unscripted conversations and voice command dictations. The dataset can be previewed here. We are in the process of publicly releasing this dataset for research and educational purpose in 2025.
Mentoring Future Leaders in Equitable Technologies
In 2024, we have hosted master degree projects and summer/all-year-long internships for over two dozen students from local high schools and prominent universities to develop their skills in UX design, HCI research, software engineering, Machine Learning, and AI with community-centric, public interest problems such as inclusive videoconferencing and fair speech AI technologies for people who stutter. Through these internship and project-based mentorship programs, we were able to broadly collaborate with educational institutions and directly engage and coach STEM students on designing and developing equitable, empowering technologies with marginalized social groups, advancing the our mission not only by the direct outcome of these projects but also the potential of the new generation of technical talents who share our values and vision with technology.
The students and teams we have worked with in 2024 include (but not limited to):
- Cornell University, Information Science MS Project Spring 2024. āDesigning Mindingful Online Meeting Experiences for and with People Who Stutterā. Aditi Ramprasad, Tammy Zhang, Yuning Zhang, Roxanne, Tanenbaum
- Public Interest Technology New England (PIT-NE), Impact Technology Summer Fellowship Project. Summer 2024. āBenchmarking Stuttered Speech Against Automatic Speech Recognition Modelsā. Dongim Lee (Olin College), Ishita Kakkar (UMass Amherst), Nikhila Peravali (UMass Amherst)
- Boston University Spark! 2024 Fall Project. āBenchmarking Chinese Stuttered Speech Recordings Against Speech Recognition Modelsā. Wai Yuan Cheng, Mary Choe, Lingjie Yuan, Yol-Emma Veillard
- University of California, Santa Cruz. UX and design internship. Summer 2024. Anusha A. Kshirsagar, Kendra Wu, Kathy Su, Huynh Yen Nhi Tran, Bridget Ho, Bianca Nicle Villamor.
- Research & engineering interns: Kexin Feng (Texas A&M University), Aishwarya Pai (Santa Clara University), Charan Sridhar (BASIS Independent Silicon Valley), Greg Price (University of Ottawa)
- Design interns: Andrew Zhang (Cornell University), Adedotun Bello (University of Pennsylvania),Ā
While all the students have made incredible contributions to our organization, we are also very proud to see their growth as designers, engineers, and researchers! Two of our interns have turned their internship work into first-authored paper submissions to top-tiered research conferences this year. Congratulations on their achievements and we look forward to seeing them all become the next generation leaders for equitable and responsible technologies!
Advocating for Community-Centric Accessibility and Inclusion
Although deeply engaged in the technology space, we know that the challenges we are dealing with are not only technological but societal. It takes not only one community or one product, but structural changes in social values, cultural practices, and public policy to enact long-term, sustainable progress.
As part of our commitment for systemic changes, AImpower.org has actively engaged in the conversation with the tech community and the public to advocate for the values and the rights of disability communities in driving the creation of technologies. We have spoke at numerous conferences, public and community events, podcasts, and university classrooms, as well as our online presence, sharing the knowledge and lived experiences with disability to create better AI systems and building a public discourse on inclusive and fair technologies:
- Panel presentation:Ā āThe Value and Value System of AIā. Online. 80 attendees.
- Panel presentation: āBlocked by the System: How Current Voice AI Silences People Who Stutterā. American Association for Advancement in Science (AAAS) Annual Meeting. Boston. 30 attendees
- Conference presentation: āFinding My Voice over Zoomā. CHI 2024. Honolulu. 40 attendees
- Poster presentation: āTowards Fair and Inclusive Speech Recognition for Stutteringā, CHI 2024. ~50 attendees
- Conference presentation: āRe-envisioning Remote Meetingsā. DIS 2024. ~50 attendees.
- CSCW conference presentation: “”, ~40 attendees
- InterSpeech conference presentation, ~100 attendees
- Guest lecture: “AI and People with Disabilities”. Stanford University CS-377Q/ME-214 Designing for Accessibility (Spring 2024)
- Guest lecture: “Fair and Inclusive Speech AI for Stuttering”. University of Ottawa DTO 5310 Ethics for Design, AI and Robotics (Winter 2024)
- Podcast: AIēé°ęé¢ļ¼ē¹ę¦®čå¾ēę®é人士å°å¢ (The Dark Side of AI: The Struggle of Disabled People Behind the Boom). ēę²¹ęē¤éŗµå (Avocado Toast). [Episode #129]
- News coverage: āRedesigning videoconferencing for, and by, people who stutterā. Cornell Chronicle, July 9, 2024
- Joint public campaign (with ~100 stuttering organizations) “@sumsung, suppressing stuttering is not the solution“.
- Tech4All bi-weekly newsletter: 8,429 views and 5,616 readers in 2024 (10x and 15x of 2023, respectively); 108 subscribers (80% growth since end of 2023)
- Website: 28K impression (282% increase from 2023), 15.3K unique visitors (427% increase from 2023)
- LinkedIn: 4000+ impression in 2024
Looking Ahead to 2025
I donāt just want us to get a seat at someone elseās table, I want us to be able to build something more magnificent than a table, together with our accomplices.
– Mia Mingus
Our progress and growth in 2024 has positioned AImpower.org as a leading voice in community-led, public-interest technologies. Moving forwards in 2025, we will continue pushing the envelop for affirmative, empathetic technology and build something magnificent with all of you:
- Expand our community engagement and partnerships;
- Publicly launch and continuously improve LibOrate;
- Advance ethical data collection and governance model for marginalized and underrepresented groups in AI;
- Create a large-scale, community-centered stuttered speech AI dataset in English;
- Co-design disfluency-friendly, stuttering-affirmative speech AI products with the stuttering community;
- Continue advocating for fair and accessible speech AI models on global platforms.
Weāre incredibly grateful to our supporters, volunteers, and community members who make this work possible. While 2025 has started with some disheartening changes in the funding and policy environment that would definitely impact our work and personal lives, we are more committed than ever to our vision of interdependence and co-liberation of all marginalized groups in creating a just and equitable technological future. We believe in the communities we have built, the technologies we co-created, and the capacities and self-determination fostered in this process — all of which will outlast one project or a political cycle. Hereās to another impactful year ahead, and your support is what keeps us forward!




