Released builds available · tool

Verbatim

Local desktop speech-to-text for everyday writing

Verbatim is a speech-to-text application based on Handy by CJ Pais. I develop this fork for everyday dictation. The local transcription workflow runs on your device after the model is downloaded. Optional AI features may send text to the provider you configure. See the releases for available builds and the repository for setup and limitations.

My contribution: Developed Verbatim as a fork of Handy by CJ Pais, retaining upstream attribution while extending the desktop product under its own identity and direction.

WHAT THE PROJECT DOES
01Local transcription after the selected model is downloaded
02Writes transcribed speech into the active text field
03Desktop downloads for Windows x64, macOS Apple Silicon and Ubuntu x64
Technical details

After the selected speech model is downloaded, the speech-to-text step runs locally on the user's device.

Model downloads and update checks use a network connection. Optional AI text processing sends the text needed for that feature to the external provider configured by the user.

Verbatim is developed as a fork of Handy by CJ Pais. Current release assets and platform limitations are documented in the repository and release page.

Speech-to-textDesktopLocal TranscriptionTauriCross-platform