The dying language now gets a model
the dying language now gets a model: meta’s speech systems understand 1,600 languages, about 500 that no ai had ever processed, and open-source voice cloning rebuilds oral languages from minutes of audio. nankai’s pipeline ties recognition, translation and synthesis into one loop.
the last fluent speaker may no longer be the archive.
Context
Meta's Omnilingual ASR, introduced Nov 10, 2025, is a suite of speech recognition models covering more than 1,600 languages, released open source. The paper says thousands of long-tail languages remain effectively unsupported by most ASR systems.
Separate research covers endangered language speech recognition pipelines, including work on Nepal's endangered languages and the ELPIS pipeline for language documentation.
The 1,600 plus languages match Meta's post. The 500 languages no AI had processed, minutes-of-audio voice cloning and Nankai's pipeline are not seen in the sources read, so unsupported here, not refuted. 'The last speaker may no longer be the archive' is the author's.
Related work
Watch next
- Look for the Nankai pipeline and the voice cloning sources.
Sources
- Meta AI, Omnilingual ASR, Nov 10, 2025ai.meta.com
- arXiv 2511.09690, Omnilingual ASRarxiv.org
- Cambridge repository, endangered languages of Nepalrepository.cam.ac.uk
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 2 October 2026 at 09:48 IST. Sources are the papers and datasets the note draws on.
View the original post ↗Embed this note
More notes
The air is now being asked to keep its own ledger
the air is now being asked to keep its own ledger: ecmwf’s aifs compo becomes the first ai model to forecast atmospheric composition globally every three hours, cleanair simulates 365 days of pm2.5 over china in ten seconds, and a unified framework maps six pollutants at one kilometer across the whole country. the air now files its own composition report.
read the note →The current is now being asked to draw its own map
the current is now being asked to draw its own map: china’s langya 2.0 predicts six ocean phenomena including internal waves and mesoscale eddies, a deep net called wenhai resolves eddies globally with air sea flux formulas built in, and scripps infers surface currents from the way temperature patterns deform in satellite images. the ocean now files its own circulation report.
read the note →The soil is now being asked to report its own carbon
the soil is now being asked to report its own carbon: a nix color sensor paired with generative data augmentation predicts soil organic carbon without a lab, random forest drives 74 percent of soil health mapping studies, and sentinel 2 tracks five year carbon change across france and italy from 922 samples. the dirt now files its own carbon account.
read the note →