Skip to main content
Language support for Real-time STT depends on the model you select. AssemblyAI offers several streaming models with different language capabilities:

​
Available Models and Language Support

ModelLanguages Supported
Universal-3.5 Pro Streaming (Preview)English, Spanish, German, French, Portuguese, Italian, Turkish, Dutch, Swedish, Norwegian, Danish, Finnish, Hindi, Vietnamese, Arabic, Hebrew, Japanese, Urdu, Chinese (native code switching)
Universal-3 Pro StreamingEnglish, Spanish, German, French, Portuguese, Italian (native code switching)
Universal-Streaming EnglishEnglish only
Universal-Streaming MultilingualEnglish, Spanish, German, French, Portuguese, Italian (per turn)

​
Choosing the Right Model

  • If you need real-time transcription in a language beyond EN/ES/DE/FR/PT/IT (for example Japanese, Chinese, Hindi, Arabic, or Hebrew), use Universal-3.5 Pro Streaming (universal-3-5-pro, Preview).
  • If you need the highest accuracy with multilingual code switching support across EN/ES/DE/FR/PT/IT, use Universal-3 Pro Streaming (u3-rt-pro).
  • If you only need English transcription at the lowest cost, use Universal-Streaming English (universal-streaming-english).
  • If you need multilingual support at a lower cost, use Universal-Streaming Multilingual (universal-streaming-multilingual).
For more details on model selection, see the Model selection page.

​
Difference From Pre-recorded STT

It’s important to note that the language_code parameter mentioned in some AssemblyAI documentation applies to the Pre-recorded STT feature, not the Streaming Transcription feature. For real-time STT, you specify the model using the speech_model parameter.
To stay informed about new features and improvements, including language support updates, you can follow our Changelog.