The State of Voice Coding
AI-assisted coding is here to stay. Products like Windsurf, Cursor, Copilot, etc. are here to stay. They provide a great, interacti...

Deepgram

Deepgram's development team experimented with voice-first coding workflows and found significant productivity improvements. Voice interfaces can transform development by enabling natural conversation with AI coding assistants, faster brainstorming sessions (3-5x speed increase), and better idea capture. The approach involves three phases: voice-first design and brainstorming, voice-enhanced code review, and selective voice-to-code implementation. While voice dictation for coding presents challenges like handling syntax and indentation, voice pairing with multimodal AI shows promise for true pair programming experiences. The transformation requires rethinking development workflows around conversational interaction rather than just adding speech recognition to existing tools.

Our dev team tried replacing typing with talking and it's working

Getting Started: A Practical Adoption Framework

Building the Future of Voice-First Development

<p>The next step will be: comic programming.
“Our team replaced the code with hand drawn comics, showing user actions on page and expected result. We are now coding way slower, but who cares, our work is now an art!”</p>


<p>New red teaming engagement - Eavesdrop employees to listen on secrets</p>
