> [Apologies for cross-posting] > > ***** Registration is now open for the Quechua ASR Shared Task @ SIMBig 2026! ***** > > ========================================================================== > Call for Participation: QUECHUA ASR SHARED TASK @ SIMBig 2026 > Automatic Speech Recognition (ASR) and Language Resources for Indigenous Languages > ========================================================================== > > Where: SIMBig 2026, Universidad Católica San Pablo, Arequipa, PERU > When: October 28 - 30, 2026 > Shared Task Registration: https://forms.gle/112jyHYPq9SVMmvs5 > Baseline & evaluation scripts: https://github.com/QuechuaBase/asr-puno-quechua > Conference website: https://simbig.org/SIMBig2026/ > > ========================================================================== > > OVERVIEW > ---------------------------------- > > SIMBig 2026 will host a Shared Task on speech technologies and resources for Indigenous languages, with an initial emphasis on Puno Quechua. We invite researchers, students, industry practitioners, and Indigenous communities to build reproducible ASR systems and open resources that support linguistic diversity, language preservation, and the South American language research ecosystem. > > > TASKS > ---------------------------------- > > Task 1: Puno Quechua ASR Baseline — Improve WER/CER on the Common Voice Puno Quechua datasets (Scripted <https://mozilladatacollective.com/datasets/cmqio1wd400xynq079ifamac6>, Spontaneous <https://mozilladatacollective.com/datasets/cmqialqsu007bnr07e4se1rl8>); baseline and evaluation scripts provided <https://github.com/QuechuaBase/asr-puno-quechua>. > Task 2: ASR for Any Indigenous Language of South America — Build an ASR system for a language that does not yet have one; teams provide their own train/dev/test splits (test ≥ 30 minutes). > Task 3: Educational/Literacy Resource — Create a resource (website, tutorial, notebook, visualisation, video, learning platform, etc.) that helps people learn, explore, or reuse Indigenous language datasets. > > AWARDS (and growing!) > ---------------------------------- > > Task 1: $100 USD + free registration to SIMBig 2026 > Task 2: $100 USD + free registration to SIMBig 2026 > Task 3: $50 USD + free registration to SIMBig 2026 > In addition, the best submission(s) will be invited to publish their work free of charge in Springer as part of the SIMBig 2026 Proceedings. Registration fees may also be waived for selected outstanding submissions. > > > IMPORTANT DATES (all times AoE) > ---------------------------------- > > 10/08/2026: Registration opens (form <https://forms.gle/112jyHYPq9SVMmvs5>) > 24/08/2026: Open orientation hour (Register <https://forms.gle/FnL2Xp8X5j3zCsHZ9>, add to calendar <https://calendar.google.com/calendar/event?action=TEMPLATE&tmeid=NnE0Ym4zajNscWxxZjVtMm5lcjdzNzhhcnIgZWx3aW5saHFAbQ&tmsrc=elwinlhq%40gmail.com>) > 16/09/2026: Open orientation hour (Register <https://forms.gle/FnL2Xp8X5j3zCsHZ9>, add to calendar <https://calendar.google.com/calendar/event?action=TEMPLATE&tmeid=MWppY2dtZ203Y2Nka2M0MTF2ZWlqamtwMDcgZWx3aW5saHFAbQ&tmsrc=elwinlhq%40gmail.com>) > 21/09/2026: Registration deadline (form <https://forms.gle/112jyHYPq9SVMmvs5>) > 30/09/2026: Release of test dataset for Task 1 > 18/10/2026: Submission deadline for final results and system description paper > 28–30/10/2026: Results announcement at SIMBig 2026 > > SUBMISSION > ---------------------------------- > > Tasks 1 & 2: a zip with one TSV per language (e.g., qxp.tsv: audio file name, predicted transcription), due 1 week after the test audio is released. Top teams will be asked to share their model and inference script for reproduction. > All tasks: a 4–8 page system description paper (Springer LNCS template <https://www.overleaf.com/latex/templates/springer-lecture-notes-in-computer-science/kzwwpvhwnvfj>) via CMT <https://cmt3.research.microsoft.com/SIMBig2026>. Papers are peer-reviewed (2+ reviewers) and plagiarism-screened; winning papers will be published in the Springer CCIS Series <https://www.springer.com/series/7899> (to be confirmed). > > ORGANISERS > ---------------------------------- > > Programme Chairs: > > Elwin Huaman, University of Cambridge, UK - CONTACT (elh97@cam.ac.uk <mailto:elh97@cam.ac.uk>) > Johanna Cordova, ERTIM at Inalco, France > Hugo Alatrista, De Vinci Research Centre, France > Juan Antonio Lossio-Ventura, National Institutes of Health, USA > Programme Committee: (Expression of Interest for reviewers <https://forms.gle/ud5kUsToLdGUexHv9>) > > Johanna Cordova, ERTIM at Inalco, France > Elwin Huaman, University of Cambridge, UK > Advisory Committee: > > Anna Korhonen, University of Cambridge, UK > Damien Nouvel, ERTIM at Inalco, France > > ========================================================================== > SIMBig 2026 — WHY JOIN US IN AREQUIPA > ========================================================================== > > The shared task is part of the 13th International Conference on Information Management and Big Data (SIMBig 2026), an international conference on AI, Data Science, Machine Learning, and NLP. Participants are invited to attend the full conference. > > > KEYNOTE SPEAKERS > ---------------------------------- >  Hugo AERTS, Harvard University, USA  Omar ALONSO, Amazon Science, USA  Gabriela CSURKA, Naver Labs, France  Dan JURAFSKY, Stanford University, USA  Tom M. MITCHELL, Carnegie Mellon University, USA  Hoifung POON, Microsoft Research, USA > > MORE DETAILS > ---------------------------------- > > Publication: all accepted papers appear in the Springer CCIS Series <https://www.springer.com/series/7899> > Travel grants: four student travel awards supported by NAACL <http://naacl.org/> > Special tracks: DISE <https://simbig.org/SIMBig2026/en/dise.html> (Data-Driven Software Engineering) and RACV <https://simbig.org/SIMBig2026/en/racvimdl.html> (Recent Advances in Computer Vision) > Co-located school: LASAI 2026 <https://simbig.org/lasai/2026/>, Latin American School of AI, October 23–26, UPC, Lima (in-person and online) > Contact: SIMBig 2026 General Chairs — Juan Antonio Lossio-Ventura (juan.lossio@nih.gov <mailto:juan.lossio@nih.gov>), Hugo Alatrista-Salas (hugo.alatrista_salas@devinci.fr <mailto:hugo.alatrista_salas@devinci.fr>)