Skip to main content
Set via transcription_engine and model query parameters.

Engines

EngineDefault modelOther modelsNotes
Deepgramnova-3nova-2, fluxDefault engine. Broadest format support.
Telnyxopenai/whisper-tinyOn-network, lightweight
Googlelatest_longMultilingual, long-form
Azureazure/fastBroad language/accent coverage
xAIxai/grok-sttGrok STT for real-time transcription
AssemblyAIassemblyai/universal-streamingUniversal-Streaming for low-latency voice agents
Speechmaticsspeechmatics/standardHigh-accuracy real-time transcription with multilingual and bilingual packs
Sonioxsoniox/stt-rt-v4Real-time transcription with automatic language detection

Flux Model

Deepgram’s lowest-latency model with built-in end-of-turn detection. Designed for real-time voice agents. See Audio Formats for supported formats.