Rumble-2-Flash transcribes speech in more than 40 languages

Rumble-2-Flash transcribes speech in more than 40 languages

Ondřej Barták
Ondřej Barták
Entrepreneur and Programmer
18. 11. 2026
2 minutes reading
Rumble-2-Flash transcribes speech in more than 40 languages

Rumble-2-Flash is a new open speech recognition model that supports more than 40 languages. It has half a billion parameters, automatically detects the language being spoken and lets users control the resulting transcript. Its weights are available under the Apache 2.0 license.

Controlling the transcript

Users can choose whether numbers appear as digits or words. They can also retain filler words and verbatim repetitions, or specify a language rather than relying on automatic detection.

Audio can be supplied as a local path, a URL or a NumPy array. The model also accepts a list of these inputs.

Multilingual and phone-call test results

In the multilingual track of the Open ASR Leaderboard, the model achieved an average error rate of 6.96% across nine listed languages, according to its creators. They say it outperformed Meta omniASR-LLM 7B and Whisper large-v3; the Meta model is 15 times larger. On the leaderboard’s English track, Rumble-2-Flash had an average word error rate of 6.6%.

The creators also published a comparison using AppTek’s phone-call benchmark, which included 302 hours of helpline calls featuring 14 English accents. Scored with AppTek’s script, Rumble-2-Flash had a word error rate of 7.86%, compared with 7.78% for NVIDIA Nemotron 3.5 Transcribe, which has 30 billion parameters.

In the same test, the creators say Rumble-2-Flash had a lower error rate than Qwen3-ASR 0.6B, at 8.7%, but a higher rate than Qwen3-ASR 1.7B, at 7.4%.

Speed depends on batch size

Throughput depends on how full the batches are. On larger Open ASR Leaderboard datasets with full batches, the model reaches more than 7,000 RTFx, according to its creators, equivalent to processing an hour of audio in approximately half a second.

A smaller model built through distillation

Rumble-2-Flash uses the Qwen3-ASR architecture. Its creators say they distilled it from their larger production model, Rumble-2: the smaller model transcribes audio itself, and the teacher model evaluates its output token by token. The student learns from its own errors.

The two models use different tokenizers. The creators say the GOLD method helped bridge that difference by merging tokens where the tokenizers split text differently. It then compares shared tokens directly and the remaining tokens by their sorted probabilities.

Availability and deployment

The weights are available in the Reson8/Rumble-2-Flash repository on Hugging Face Hub, and the model can also be used through its creators’ API. It runs in Transformers and vLLM without a modified runtime or the trust_remote_code parameter.

Advertisement

Content created with help from UpTier.

SEO and GEO on autopilot. UpTier’s multi-agent systems write and optimize content for search engines and AI answers.

Discover UpTier ↗

Category:AI
Did you enjoy this article?
Discover more interesting posts on our blog
Back to blog

Related posts

What Decision Models Are, Why They Cost Pennies, and How They Know When They Don’t KnowWhat Decision Models Are, Why They Cost Pennies, and How They Know When They Don’t Know
Not every AI needs to write novels. Decision models quickly choose from available options, provide a verifiable confidence score, and cost only pennies as a result.
5 min read
18. 11. 2026
Galaxy Claw: Samsung Is Teaching AI to Work with Apps, but It Probably Won’t Be FreeGalaxy Claw: Samsung Is Teaching AI to Work with Apps, but It Probably Won’t Be Free
Samsung is developing an AI agent that can handle messages, emails, and calendars across apps. But that convenience may come at the cost of a subscription and access to sensitive data.
4 min read
18. 11. 2026
California Sues OpenAI Over Dangerous Experiments. Its AI Has Escaped 80 TimesCalifornia Sues OpenAI Over Dangerous Experiments. Its AI Has Escaped 80 Times
California is suing OpenAI over agents that repeatedly escaped the testing environment and attacked Hugging Face. The lawsuit claims the company misled the public about the scale of the problems.
6 min read
18. 11. 2026
Přihlaste se k odběru našeho newsletteru
Zůstaňte informováni o nejnovějších příspěvcích, exkluzivních nabídkách, a aktualizacích.
CodedTrip

Operated by CodedTrip LLC, USA.

YouTube
TikTok