Deepgram
Real-time speech-to-text API for developers
A solid emerging choice. Best if you already know you need audio and want to experiment.
Deepgram provides low-latency speech-to-text, text-to-speech and voice agent APIs (Nova-3, Aura) with strong accuracy on noisy calls and streaming audio.
Building production real-time transcription, voice agents and call analytics.
Knowledge workers across functions
Casual or one-off users who won't use it weekly
Pros & cons
- Streaming STT
- Strong fit for building production real-time transcription, voice agents and call analytics.
- Transparent, published pricing
- Advanced features may require a higher tier
- Output quality varies with prompt skill
- Limited offline use — internet required
Why we scored it 64/100
The ToolHund Score blends five signals. Hover any bar for the weight. See the full rubric →
We refresh entries monthly and re-score whenever pricing, features, or reviews change materially. See methodology →
Steep — expect a few hours to feel productive
Great for small teams (2–20) and serious solo pros
Positive once you replace 1+ hour of manual work per week
What you get
- Streaming STT
- Nova-3 model
- Aura TTS
- Voice Agent API
Compare Deepgram with alternatives
Pick a rival and see the side-by-side breakdown, plus how the community votes.
More in Audio
Community reviews
Why trust ToolHund?
ToolHund is curated, not crawled. We score tools by real-world fit, feature depth, pricing transparency, community signal and shipping momentum. Sponsored placement never changes the ToolHund Score.