en

Supported Languages in Transcription

Supported transcription languages: over 160 languages and variants, automatic detection, translation for ~100 languages, and multi-language document export.

VoxScriber transcribes in over 160 languages and variants, with Brazilian Portuguese and European Portuguese at the top of the priority list. You choose the audio language before transcribing — or let automatic detection decide.

How to choose the language

On the new transcription screen, select the audio language from the list. Two tips:

  • Know the language? Select it. Entering the correct language gives the best result.
  • Don't know or is it variable? Use "Automatic detection". The automatic option is also available for URL/YouTube import and on the Nano engine (browser), where it's the default.

Each transcription engine uses its own language codes — the platform handles that conversion for you, so you only choose the language once.

For Portuguese audio with technical terms (proper names, medications, legal jargon), the Premium engine accepts contextual reinforcement terms — which greatly improves accuracy for these words. See Accuracy and how to improve.

And after transcription? Translation and documents

Language doesn't end with transcription:

  • Transcription translation: the finished text can be translated into about 100 languages, with the result stored per language (translating again doesn't cost another quota). The monthly translation quota varies by plan.
  • Summary, title, and minutes: AI features respond in your interface language — if you use the site in Portuguese, the summary comes out in Portuguese, even if the audio is in English.
  • Minutes in PDF: can be generated in 17 languages;
  • PDF export: the document is formatted in your language (Portuguese, English, French, Spanish, German, Italian, Japanese, and Arabic, among others).

Subtitles in other languages

SRT and VTT formats are generated from the transcription — in the language in which it was made. For subtitles in another language, transcribe first and then use transcription translation.

Frequently asked questions

Does automatic detection work well?

Yes, especially for audio with a dominant language and good quality. If you already know the language, selecting it manually remains the most accurate.

What if the audio has two mixed languages?

Choose the predominant language or use automatic detection. There's a dedicated article: Multiple languages in the same audio.

Do regional accents interfere?

The Standard engine is the most robust for heavy accents and difficult recordings; Premium is optimized for Brazilian Portuguese. It's worth testing both on your type of audio.

The summary came out in Portuguese, but the audio was in English. Why?

AI features follow the interface language, not the audio language. Change the site language if you want the summary in another language.


Still have questions? Open a ticket — normal-priority tickets are answered within 24 hours.