Polyglot voice intelligence for 98 languages

Upload or record audio, choose the processing model, and turn voice into multilingual text within seconds.

Accurate transcription

High-quality Whisper-based speech-to-text in 98 languages with smart corrections.

Real-time capture

Record microphone, headphones, or system audio with live progress and minute tracking.

Community and support

Collaborative chat, proposals board, and moderation tools that keep the platform safe.

98 languages

Polyglot Voice Geography

The more popular the language, the higher the base confidence of the model. Rare languages are supported, but processing may take a bit longer.

Available now

98

Chinese (Simplified)

zh

English

en

French

fr

German

de

Japanese

ja

Portuguese

pt

Russian

ru

Spanish

es

Afrikaans

af

Albanian

sq

Amharic

am

Arabic

ar

Armenian

hy

Assamese

as

Azerbaijani

az

Bashkir

ba

Basque

eu

Belarusian

be

Bengali

bn

Bosnian

bs

Breton

br

Bulgarian

bg

Burmese

my

Catalan

ca

Technical requirements

Audio upload and recording

  • Formats: mp3, wav, m4a, mp4, mpeg, mpga, webm. Recommended .mp3, .wav or .m4a — they are the most stable.
  • File size: up to 25 MB. If exceeded, the upload stops automatically.
  • Recording duration: up to 5 minutes on the free plan. The timer stops automatically.
  • Recognition quality: Popular languages (top-20) are processed with maximum confidence. For rare languages, results depend on audio clarity.
  • Free quotas: 3 attempts per model and 5 minutes of recording are available immediately after registration and do not expire.

Tip: if audio is longer than 5 minutes, split it into parts or use paid recording minutes to avoid trimming.

Partner ads