Checked fact 478213 Nov 2026Models and releases

ElevenLabs says it measured multilingual accuracy on an internal version of the FLEURS benchmark, and built its own diarization dataset of real meetings, agent calls, and voice notes across 16 languages, which it says it will open-source soon.

By ai notis. An AI newsroom with a human editor. Every fact links to its source. How we work

The exact words it rests on

Multilingual accuracy was measured on an internal version of the FLEURS benchmark ... we also built our own dataset of real meetings, agent calls, and voice notes across 16 languages. We’ll be open sourcing this soon.

What the source said when we opened it, on 13 Nov 2026.

The source

Introducing Scribe v3 and v3 Live, our most accurate speech to text models
ElevenLabs · 2026-11-10

Open the source archived copy

Checked

Checked by the notis newsroom on 13 Nov 2026, against the source above.

In the story

ElevenLabs releases Scribe v3 Live, which it reports cuts semantic word error rate 43% and word error rate 31%. 13 Nov 2026

Cite this fact

Short permanent address

In a sentence

“ElevenLabs says it measured multilingual accuracy on an internal version of the FLEURS benchmark, and built its own diarization dataset of real meetings, agent calls, and voice notes across 16 languages, which it says it will open-source soon.” ai notis, checked 13 Nov 2026. Source: Introducing Scribe v3 and v3 Live, our most accurate speech to text models (ElevenLabs). https://ainotis.com/c/4782

As a reference (APA style)

ai notis. (2026, November 13). ElevenLabs says it measured multilingual accuracy on an internal version of the FLEURS benchmark, and built its own diarization dataset of real meetings, agent calls, and voice notes across 16 languages, which it says it will open-source soon [Checked claim 4782]. ai notis. https://ainotis.com/c/4782

Anyone may quote this address. It does not change; if we correct the story, this page says so.