The latest articles on Forem by Harsh Sharma (@zherotz). https://forem.com/zherotz 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%2F3777119%2F07722933-15f9-4b2d-a003-c02453130e5d.png https://forem.com/zherotz en Harsh Sharma Tue, 17 Feb 2026 07:55:33 +0000 https://forem.com/zherotz/built-toice-local-voice-typing-for-linux-1gnk https://forem.com/zherotz/built-toice-local-voice-typing-for-linux-1gnk <p>I built a simple voice-to-text app for Linux because existing solutions either don't work well or send your audio to the cloud.</p> <h2> What it does: </h2> <ul> <li>Press hotkey → pill popup appears</li> <li>Speak your text</li> <li>Stop speaking → Ctrl+V anywhere to paste</li> </ul> <h2> Why it's different: </h2> <ul> <li>100% local using Whisper models</li> <li>No cloud services, no API costs</li> <li>Works system-wide (terminal, browser, IDE, everywhere)</li> <li>You choose the model size (fast vs accurate)</li> </ul> <p>Built it for myself on Fedora KDE, but should work on most distros.</p> <p>Check it out: <a href="https://github.com/harshsharma2455/Toice" rel="noopener noreferrer">https://github.com/harshsharma2455/Toice</a></p> <p>Open to feedback and contributions!</p> linux opensource ai privacy