Revision History
Automatic Speech Recognition Models · 11 revisions
Sizes are character counts of the article source. The signed number is the change from the previous revision.
Recent edit summaries
Detailed summaries recorded by editors. Generic maintenance summaries are omitted here; every recorded revision remains below. Dates describe the edit, not necessarily the event it covers.
- Correction: Corrected the release dates of Deepgram Nova-2, Nvidia Parakeet-TDT 1.1B and IBM Granite Speech 3.3, the Distil-Whisper parameter label, the VoxPopuli transcribed-hours figure, an unsupported 32 percent Universal-2 claim, and the institution behind the SUMMIT recognizer (plus 19 minor fixes)
Version 11 · Nov 5, 2026, 11:53 AM
- Update: add Nemotron 3.5 Transcribe (28 Oct 2026) to the ASR systems table
Version 10 · Oct 29, 2026, 09:08 AM
- Correction: reference 23 author garbled as 'Communication Communication'; corrected to the arXiv listing (Seamless Communication, Barrault, Chung, Meglioli, Dale, et al.) and exact title
Version 9 · Oct 16, 2026, 11:26 PM
- Added Meta Omnilingual ASR (Nov 2025) to the production systems table with sourced figures and a link
Version 8 · Oct 16, 2026, 11:06 PM
- Improve internal links (2026-10-13)
Version 7 · Oct 13, 2026, 10:01 AM