Updated

AssemblyAI-STT-3.7-Realtime

AssemblyAI released AssemblyAI-STT-3.7-Realtime for live conversations, including meetings, calls, and captions. It supports 32 languages in one stream with automatic detection and real-time speaker labels.

By No Code MBA · Facts checked

Provider
AssemblyAI
Announcement date
Status at last check
Generally available

What changed

AssemblyAI-STT-3.7-Realtime adds live speaker labels finalized at turn end and supports 32 languages in one stream with automatic detection. AssemblyAI says it replaces Universal-Streaming and Universal-Streaming Multilingual.

Access and availability

Use the AssemblyAI Realtime Speech-to-Text API and set speech_model to assemblyai-stt-3-7-realtime.

Official access documentation

What you can build

Try it on live audio where speaker labels need to appear during the session, and check whether its transcript-focused turn segmentation fits the interaction.

Limitations

AssemblyAI says this model is not tuned for voice-agent turn-taking. Its launch announcement positions the Realtime-Agent variant for that use case.

Sources and release history