AssemblyAI provides high-accuracy cloud-based speech recognition with real-time WebSocket streaming (v3 protocol) and an automatic REST fallback. The plugin supports two model tiers: Universal-3 Pro for top accuracy in 6 major languages, and Universal-2 for broader coverage across 37 languages.
Features
Real-time WebSocket streaming with interim and final results
Automatic REST fallback if streaming is unavailable
Language detection
Smart formatting
API key validation in settings
Transcription Models
Model
ID
Languages
Universal-3 Pro
universal-3-pro
6 (de, en, es, fr, it, pt)
Universal-2
universal-2
37
Configuration
API Key - Create an API key at assemblyai.com. Requires an account.
Setup
Open TypeWhisper Settings > Plugins
Find the AssemblyAI plugin and click Configure
Enter your AssemblyAI API key (validated automatically)
Select your preferred model
Select AssemblyAI as your transcription engine in Settings or in a profile
