How to read this guide
This article is more of a reference, rather than a blog. Use it in the same way you’d use a dictionary or encyclopedia.

Skim the TL;D...

Deepgram

Deepgram offers two Nova speech-to-text models optimized for different use cases. Nova-2 prioritizes speed and cost-effectiveness for English-heavy batch processing, while Nova-3 excels in real-time multilingual transcription with advanced features like keyterm prompting. The guide provides detailed benchmarks, cost comparisons, and implementation recommendations to help developers choose the right model based on accuracy requirements, latency constraints, language support needs, and budget considerations.

Model Comparison: When to Use Nova‑2 vs Nova‑3 (for Devs)

Why Model Choice Matters in Choosing Speech-to-Text (STT) APIs for Your Apps

Which Model Is Right for You? Nova-2 or Nova-3?

Benchmarks and Performance Tests That Matter

Production Checklist for Devs Shipping Deepgram’s Speech-to-Text (STT) Models

Next Steps: When to Use Nova-2 vs Nova-3 (for devs)