Submit your tool

Built for devs.

Twitter Submit a tool Share feedback

Built for devs.

Twitter•Submit a tool•Feedback

Deepgram

Fast and accurate speech-to-text and text-to-speech APIs

Free ($200 credit) | PAYG +1Voice AILast verified: 2026-03-09

Overview

Deepgram provides industry-leading speech-to-text (STT) and text-to-speech (TTS) APIs with sub-300ms latency for real-time applications. Nova-3 model achieves 90%+ accuracy and transcribes an hour of audio in about 12 seconds. Supports real-time streaming and pre-recorded audio with per-second billing. Features include speaker diarization, language detection, summarization, sentiment analysis, and redaction. TTS with Aura-2 offers 40+ voices in 7 languages. Voice Agent API bundles STT and TTS starting at $0.05-0.08/min for conversational AI. Start free with $200 credit. SOC 2, HIPAA compatible, GDPR compliant.

Works with

REST API

Pricing

$0 + usagePay-As-You-Go

$200 free credits
Nova-3 STT: $0.0043/min pre-recorded
Nova-3 STT: $0.0077/min streaming
Aura-2 TTS: $0.030/1k chars
Up to 100 REST / 150 WSS concurrency
Community and Discord support

$4,000/yr prepaidGrowth

~15-20% discount on PAYG
Nova-3 STT: $0.0036/min pre-recorded
Nova-3 STT: $0.0065/min streaming
Aura-2 TTS: $0.027/1k chars
Higher concurrency limits

CustomEnterprise

Volume discounts
On-premise deployment
Custom model training
99.9% SLA
Dedicated support

Pros

+Industry-leading speed and accuracy
+$200 free credits to start
+Per-second billing
+Voice Agent API for conversational AI

Cons

-Add-ons like diarization cost extra
-Enterprise pricing not transparent

Find similar tools

ElevenLabs

Most realistic AI voice generation and cloning

ElevenLabs

Most realistic AI voice generation and cloning

Get Discovered by Developers

Promote your tool

Reach thousands of developers actively searching for AI tools. Featured listings get 10x more clicks.