Deepgram
Fast, accurate speech-to-text API
About
Real-time and batch speech recognition API with <300ms latency. Supports 30+ languages, speaker diarization, and custom vocabulary. Nova-3 model is best-in-class for English accuracy.
Choose Deepgram when…
- •You need real-time speech-to-text with <300ms latency
- •Accuracy on noisy audio or accented speech matters
- •You need speaker diarization or custom vocabulary
Builder Slot
Speech synthesis and recognition APIs — text-to-speech, speech-to-text, and real-time audio intelligence
Other tools in this slot:
Stack Genome Detection
AIchitect's Genome scanner detects Deepgram in your project via these signals:
@deepgram/sdkdeepgram-sdkDEEPGRAM_API_KEYIntegrates with (4)
Deepgram STT API is wrapped as a LangGraph tool node, transcribing audio at agent decision points.
→ Process voice input inline in a LangGraph agent — no separate audio pipeline required.
Vapi uses Deepgram as its real-time speech-to-text engine for caller transcription.
→ Fast accurate caller transcription in Vapi — Deepgram STT feeds context directly to the agent.
Pipecat ships a Deepgram STT service that streams audio frames in and transcribed text out as a pipeline node.
→ Drop Deepgram STT into a Pipecat voice agent pipeline with a single service binding.
LiveKit Agents ships a first-party Deepgram STT plugin that streams room audio in and transcribed text out.
→ Wire Deepgram STT into a LiveKit voice agent with a single plugin import.
Alternatives to consider (1)
Pricing
Recent Activity
View all activity for this tool →In 2 stacks
Badge
Add to your GitHub README
[](https://www.aichitect.dev/tool/deepgram)Explore the full AI landscape
See how Deepgram fits into the bigger picture — browse all 207 tools and their relationships.