Otter.ai just announced its biggest language expansion yet, adding 27 languages and a multilingual transcription feature that follows a meeting even when speakers change languages mid-call. Thirteen of the new languages work today, and the other 14 are on the way. If your team talks across borders, this changes what an AI notetaker can do for you.
Until now, Otter handled only four languages: English, French, Spanish and Japanese. Once the rollout finishes, the total will reach 31.
Otter.ai Multilingual Transcription Launch
The company is making two changes at once. First, it is adding many new languages for transcription. Second, it is building a multilingual mode that listens for language switches and keeps transcribing without anyone touching a setting.
Otter gives a simple example. A sales call might start in English, move to Portuguese, then shift to Mandarin and back, and the transcript keeps up the whole time. The language-switching feature is listed as coming soon, so it is not live for everyone yet.
Languages available now
The 13 languages live today are Mandarin Chinese (Simplified), German, Italian, Cantonese, Korean, Portuguese (Brazil), Russian, Indonesian, Malay, Dutch, Turkish, Vietnamese and Polish. You pick a supported language in your Otter settings. The company says the new languages cost nothing extra on any plan and work on desktop, mobile and the browser.
Languages coming soon
Fourteen more are planned: Swedish, Danish, Finnish, Arabic, Hindi, Thai, Hebrew, Norwegian, Greek, Macedonian, Persian, Czech, Hungarian and Romanian. Otter has not given dates for them.
How Multilingual Transcription Works in Otter
Otter says it trains and runs its own speech models on its own systems instead of wrapping a third-party AI service. The company argues this lets it tune each language properly and keep audio away from outside processors. In English tests it reports about 96 percent average accuracy on the Open ASR Leaderboard, which is a company claim we could not check.
The new language models were trained on publicly available datasets, the company told reporters. Native-speaking volunteers then reviewed transcripts to check accuracy. That is a reasonable approach, but accuracy for less common languages is something you should test yourself before relying on it.
The languages also work across the rest of the product. Summaries, action items, speaker labels, search and AI Chat are meant to function in each supported language. A newer feature called Work Mode builds a pre-meeting brief from past conversations and connected apps, then drafts follow-up emails and CRM updates in the language the meeting used.
| Item | Before | After rollout |
|---|---|---|
| Supported languages | 4 | 31 |
| New languages live today | 0 | 13 |
| New languages coming soon | 0 | 14 |
| Switching languages mid-meeting | No | Coming soon |
| Extra cost | Not applicable | None, per the company |
Why Multilingual AI Notetakers Matter
Many global teams do not stick to one language. Otter points to a survey in which 42 percent of international knowledge workers named language barriers as a top collaboration challenge. A notetaker that only understands English leaves those teams typing notes by hand.
The move also shows how crowded the meeting notes space has become. Otter has moved beyond plain transcripts into briefings, live coaching and follow-up drafts. Language coverage is one more way to stand out, and it follows the company’s large funding round from General Catalyst reported earlier this year.
What This Means for You
- If your team uses any of the 13 live languages, switch the language in Otter settings and run a short test meeting.
- If you need a language from the coming-soon list, wait for the official release before planning around it.
- For mixed-language calls, hold off on switching until the multilingual option actually ships.
- Check transcripts for names and technical terms, since accuracy claims come from the company.
- Compare it with other AI meeting tools if language coverage is your main need.
Sources
Primary source: Otter.ai’s announcement. Additional reporting from TechCrunch.
Conclusion
Otter.ai is going from four languages to 31, with 13 live now and a mid-call language-switching mode still to come. The accuracy numbers are the company’s own, so test your language before you depend on it. For international teams, though, this removes a long-standing limit on AI meeting notes and puts pressure on rivals to match it.
