Inclusive Voices: Advancing Language Technology for People with Impaired Speech in Local Languages

In April 2025, the CDLI (Centre for Digital Language exclusion) team attended CHI 2025 - the ACM (Association for Computing Machinery) conference on Human Factors in Computing Systems held in Yokohama, Japan. They took part in a workshop on the theme of "Speech AI for All: Promoting Accessibility, Fairness, Inclusivity, and Equity" and made a poster presentation on “Inclusive Voices: Advancing Language Technology for People with Impaired Speech in Local Languages”, which addressed the question of how to promote accessibility, fairness, inclusivity and equity through the use of language technology in local languages for people with impaired speech.

 

There is a need to support and develop automated speech recognition (ASR) models based on African languages, and to support the innovation of apps and other tools that can be used in everyday conversation to aid communication for people living with impaired speech.

 

While English language ASR models exist for interpreting impaired speech, nobody (to our knowledge) has addressed language models for African languages, until now. The Centre for Digital Language Inclusion (CDLI) was established to address this gap by creating technologies that support individuals with atypical speech in local languages and cultures, starting with ten African languages. There are significant barriers to developing ASR for impaired speech in Low Resource Languages (LRLs), primarily due to the lack of recorded speech samples. Existing datasets are almost exclusively in American English, with very limited representation of other languages, and even fewer from LRLs. English-focused models often do not reflect how English is spoken in Africa. To overcome these challenges, CDLI adopts a community-led, user-centric research practice, involving partnering with local institutions to collect recordings of impaired speech; developing open-source tools for data collection and ASR model building; and providing technical training.

 

CDLIs key principle is to democratise speech recognition technology by empowering local communities to create their own datasets and AI models. CDLI's work in Ghana with the Akan language serves as a pilot study towards this goal. The longer-term goal is to foster local, autonomous, and sustainable skills for creating inclusive ASR technologies that meet the specific needs of atypical speakers across Africa, and beyond.