Programme Chair of the SIMBig 2026 shared task on speech recognition and language resources for Indigenous languages of South America, starting with Puno Quechua.
A community-centred AI framework built on QuechuaBase, an open infrastructure for modelling, curating and deploying oral-first language technologies.
Presenting QuechuaBase.org, an open, community-curated infrastructure for oral-first language technology for Quechua languages and Puno Quechua.
We present the largest speech corpus for a single Quechua variety, collected with the community, and the first ASR benchmark for Puno Quechua.
We detail the integration of the Quechua language family into Common Voice, an open-source initiative to build diverse speech datasets.
We bring the Quechua language into Wikidata. We outline a technical pipeline for data curation and deployment, transforming Quechua linguistic resources into structured, …
Recording Puno Quechua voices across generations and making audio recordings available to all on Mozilla Common Voice to support language preservation, education, and research.
![WikiAndes [Bringing Wiki Projects to the Andes] featured image](/project/wikiandes/featured_hu_99f5456b04ca1995.webp)
Bringing Wiki Projects to the Andes aims to impart key knowledge onto learners and society at large.
Knowledge Graph-enabled conversational agents, University of Innsbruck and other partners, Private funding.
CHIA’s Meet the Researcher series features my work with Quechua communities in Puno, Peru, on community-led speech data, speech-first language technology and what comes next.
The results of the SIMBig 2026 shared task on speech recognition and language resources for Indigenous languages of South America, announced in Arequipa on 30 October 2026, and what they tell us about Puno Quechua ASR.
Why we organised a shared task on speech recognition and language resources for Indigenous languages of South America, what we asked teams to build, and what happens next at SIMBig 2026 in Arequipa.
