Sonora

Overview

I built Sonora, a Windows desktop app that lets you create and transcribe speech entirely offline. It combines text-to-speech, speech-to-text, and global voice typing into one local studio.

Challenge

Many speech tools send your data to the cloud, risking privacy and requiring internet. I wanted to keep everything on your PC with no external servers involved.

What I shipped

• A native app with a sleek studio UI for managing voices and dictation
• Global dictation that types into any app using a hotkey and live streaming
• First-run setup that installs a private Python runtime and downloads speech models
• Local multi-engine speech processing with offline TTS and STT
• Automatic signed updates and a public landing site to showcase the product

Result

Sonora lets users work with speech privately and securely on Windows. It supports natural voice creation, transcription, and dictation without cloud dependencies.