About this tool
Deepgram is an enterprise voice AI platform whose core capabilities include speech-to-text, text-to-speech, and voice agents.
Core capabilities
- Speech-to-Text: accurate speech recognition for call transcription, real-time analytics, and customer support.
- Text-to-Speech: Aura, Deepgram's inaugural text-to-speech model, produces natural, human-like voices to support real-time conversational AI applications and AI agents.
- Audio Intelligence: automatically extract topics and context, recognize speaker intent, analyze caller sentiment, and precisely summarize interactions.
- Voice Agents: supports real-time conversational AI applications and AI agent scenarios.
Typical scenarios
- Agent assist and call coaching
- Medical transcription
- Audio intelligence such as topic detection and sentiment analysis
- Call transcription and real-time analytics
- Voice agents and conversational AI
Who
it is for Deepgram targets enterprises, contact centers, healthcare, and developers. It offers an API and supports self-hosting.
Features & use cases
Speech-to-TextText-to-Speech (Aura)Audio Intelligence: topic detection, sentiment analysis, intent recognition, summarizationVoice agent supportAPI availableSelf-hosting supported
Quick facts
Public API available
Self-hosting supported
At a glance
Pricing
Pricing
Paid
from
from $1/yr
Pricing verified Oct 20, 2026Official pricing page
Resources
Sources
Every profile keeps its sources and latest verification date.Similar tools
Descript
descript.com
83
Descript is a tool for editing video and audio by editing text, with built-in AI features for transcription, voice cloning, and media generation.
Freemium · $16+/mo
LiveKit Documentation | The platform for voice, video, and physical AI
docs.livekit.io
81
The platform for voice, video, and physical AI agents.
