# AssemblyAI's Universal 3.6 Pro speech model is live on Vapi

Vapi has made AssemblyAI's Universal 3.6 Pro model available on its platform, introducing mid-sentence code-switching across 14 languages and reduced transcription errors for voice agents.

Language: en
Time zone: America/Los_Angeles
Current time (UTC, cached as of): 2026-11-22T13:08:00.000Z
Query date (America/Los_Angeles): 2026-11-22

HTML: https://didcodexreset.com/news/a32eefefcb5045cba30ca332.html

Content language: en
Localization state: sameLanguage

Source: Vapi · Published 10/5/2026, 11:54:24

Vapi has integrated AssemblyAI's Universal 3.6 Pro speech model into its platform. According to Vapi, the model cuts background speech in transcripts by 30% and makes 45% fewer incorrect yes-or-no confirmations.

The model also holds conversational turns until complete phone numbers, verification codes, or email addresses are gathered, and adds support for 14 new languages with mid-sentence code-switching. Users can configure their agents with the new model directly in the [Vapi dashboard](https://dashboard.vapi.ai/).

Tags: AssemblyAI, Vapi, Speech Recognition, Voice Agents

[View original post](https://x.com/Vapi_AI/status/2107146305701040233)
