Deepgram Nova-3
Transcription by Deepgram
DeepgramReady to build with Deepgram Nova-3?
Get Started FreeParameters & options
Label each utterance with a speaker index when multiple people talk.
Add per-word start/end times and confidence inside each transcript segment. Increases the size of the result considerably.
Detect the spoken language instead of assuming English. Leave off for known-English audio.
Explore similar models
OpenAIGPT-4o Transcribe Diarize
ElevenLabsScribe v2
Scribe v2 is a speech-to-text transcription model from ElevenLabs that supports speaker identification.
X.aiGrok STT
QwenQwen3 ASR 1.7B
Qwen3 ASR 1.7B is an open-source multilingual speech-to-text model developed by Qwen with 1.7 billion parameters.
MistralVoxtral Mini 3B
Voxtral Mini 3B is a 3-billion-parameter open-source speech-to-text model released by Mistral in July 2025.
CartesiaCartesia Sonic 3
Start building with Deepgram Nova-3
No API keys required. Create AI-powered workflows with Deepgram Nova-3 in minutes — free.