The latest articles on DEV Community by Shawn (@shawn_dsza). https://dev.to/shawn_dsza https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F28491%2F74b04852-f620-4ccf-a143-ede0fc6b2785.jpeg https://dev.to/shawn_dsza en Shawn Fri, 30 Jan 2026 23:37:08 +0000 https://dev.to/shawn_dsza/voxtube-convert-youtube-videos-to-audio-with-local-tts-1ae1 https://dev.to/shawn_dsza/voxtube-convert-youtube-videos-to-audio-with-local-tts-1ae1 <p>I built a tool to convert YouTube videos into podcasts</p> <h3> Body </h3> <p><strong>Problem:</strong> I kept queuing YouTube tutorials and talks but never watching them. Video demands attention in a way that audio doesn't.</p> <p><strong>Solution:</strong> VoxTube extracts transcripts from YouTube videos and converts them to audio using high-quality TTS.</p> <p>Now I "watch" YouTube during my commute, while cooking, and during workouts.</p> <p><strong>Technical details:</strong></p> <ul> <li>Built with Bun + Hono (~300 lines)</li> <li>Uses Kokoro TTS (runs locally via Docker)</li> <li>Caches generated audio</li> <li>No cloud dependencies</li> </ul> <p><strong>What I learned:</strong></p> <ul> <li>Bun's file APIs are really nice for streaming audio</li> <li>Modern TTS (Kokoro) sounds surprisingly natural</li> <li>Most YouTube videos have transcripts available</li> </ul> <p><strong>Stats:</strong></p> <ul> <li>2 weeks to MVP</li> <li>~300 lines of code</li> <li>0 monthly costs (runs locally)</li> </ul> <p>GitHub: <a href="https://github.com/shawn-dsz/voxtube" rel="noopener noreferrer">https://github.com/shawn-dsz/voxtube</a></p> <p>Happy to answer questions about the build!</p> ai vibecoding sideprojects showdev