Spoken

Two days ago, I had a shoulder operation. I’m currently working with one arm, and almost immediately something became very clear: modern computing assumes full mobility. Windows has a built-in utility and there are many other tools available, but I wanted to build my own so that I can customize it and have an extremely efficient flow. Typing, switching apps, taking notes, replying to messages - all of it becomes friction-heavy the moment one hand is out of the equation. I still wanted to work, think, and create - I just didn’t want the keyboard to be the bottleneck. That’s what led me to build Spoken. Why I Built Spoken Spoken is a lightweight voice-to-text application for Windows. You speak, it transcribes, and it types directly at your cursor - in any application. No copy-paste. No special editors. No workflow changes. You focus the app you’re already using, press a button, and start speaking. How It Works The goal was simplicity, especially while recovering: Continuous mode – pause naturally and it transcribes automatically Manual mode – speak freely and transcribe when you’re ready Model selection – choose and download the transcription model that fits your needs Article content Built one-handed. Used hands-free It works anywhere your cursor is active - editors, browsers, chat apps, terminals. Built Out of Necessity (and Used Immediately) This wasn’t a “someday” idea. I built and started using Spoken during recovery, because I needed a way to stay productive without forcing myself into uncomfortable or inefficient workflows. And something interesting happened: even with one arm, work felt smoother than before. Speaking ideas out loud is often faster than typing them. Less context switching. Less friction. More focus. The Bigger Lesson A lot of the best tools don’t start as business ideas - they start as solutions to real constraints. Accessibility-driven tools often end up benefiting everyone: When typing is painful When hands are busy When you just want ideas to flow without interruption Voice becomes a first-class input method. What’s Next Spoken is still evolving, and I’m actively using it as I build it. There’s plenty of room to improve accuracy, workflows, and deeper Windows integration - and real-world usage is shaping every decision. If you’re interested in: Voice-driven productivity Accessibility-first software Reducing friction between ideas and execution I’d love to hear your thoughts. Sometimes the best products don’t come from inspiration - they come from limitation. Check out the video: https://youtu.be/jdyZ8i_S_-k