SIMBig 2026 · Arequipa, Peru · 28–30 October 2026
Automatic Speech Recognition (ASR) and Language Resources for Indigenous Languages of South America
Submissions are closed. The results were announced at SIMBig 2026; see the results page. This page summarises the call. The official version is on simbig.org.
This shared task promotes speech technologies and open resources for Indigenous languages of South America, starting with Puno Quechua. It encourages reproducible, open research and collaboration with Indigenous language communities.
Build an ASR system using the Mozilla Common Voice Puno Quechua datasets (scripted and spontaneous speech). Baseline models and evaluation scripts are provided. Systems are ranked by WER on a held-out test set, and must be reproducible.
Develop an ASR system for an Indigenous or native South American language that does not yet have one. Teams must share their train/dev/test splits, and the test set must be at least 30 minutes long.
Create a resource, such as a website, tutorial, notebook, visualisation, video or learning platform, that helps people learn or explore an Indigenous language, or reuse its datasets.
All deadlines are Anywhere on Earth (AoE).
| Date | Milestone |
|---|---|
| Registration opens | |
| Datasets released; orientation hour | |
| Orientation hour | |
| Registration deadline | |
| Task 1 test set released | |
| Initial abstract deadline (CMT) | |
| Final transcriptions (Task 1), system description paper and video due | |
| Results announced at SIMBig 2026, Arequipa, Peru |
| Item | Tasks | Details |
|---|---|---|
| Abstract | All | Short preliminary description of the approach, via CMT, due 11 October. Could be revised until the final deadline. |
| Transcriptions | Task 1 | Predicted transcriptions of the test set as a TSV file, via the submission form (see evaluation). |
| System description paper | All | 4–8 pages in the Springer LNCS template, via CMT. It could be an expanded version of the abstract. |
| Video presentation | All | 15-minute presentation of the paper, via the submission form. |
Each paper was evaluated by at least two independent reviewers on originality, technical quality, significance and relevance. All manuscripts were screened with plagiarism detection software. Papers from the winning teams will be published, after revision in response to reviewer feedback, in the Springer Communications in Computer and Information Science (CCIS) series.
Winning teams also receive free registration to SIMBig 2026. See the award winners.
Programme Chairs: Elwin Huaman (University of Cambridge, UK), Johanna Cordova (ERTIM, Inalco, France),
Juan Antonio Lossio-Ventura (National Institutes of Health, USA), Hugo Alatrista-Salas (De Vinci Research Centre, France).
Programme Committee: Johanna Cordova, Elwin Huaman.
Advisory Committee: Anna Korhonen (University of Cambridge, UK), Damien Nouvel (ERTIM, Inalco, France).