Skip to content
UsecaseBoard
← All use cases
githubDictation

Dictate into the focused Mac app with local Whisper

At a glance

VoxFlow Pro's creator describes a hold-to-talk workflow that inserts locally transcribed speech into the active application. This workflow is listed under Dictation, Accessibility, Local AI.

SourceBuilding a tool for these workflows? See weekly rates.

What they did

1) Have macOS 14 or later, a microphone, and a text application ready. Download VoxFlow Pro from the creator's linked download page; a SHA-256 checksum is supplied there. 2) Launch it while online so it can download its Whisper model once; the default model is small. Subsequent transcription runs locally through WhisperKit. 3) Focus the text field where the words should appear. Hold a supported push-to-talk key, such as Globe/Fn, Caps Lock, or right Shift, and speak. 4) Release the key to finish the recording. The app transcribes it and types the result into the application that currently has focus. Check that inserted text before using it in your document or message.

Worth knowing

The post distinguishes the public 1.0 download from development code: GPU fallback and a stall watchdog were not yet included in that public build. The creator mainly tests English and Slovak.

Save ideas in this browser. Upvotes count across the board.

Prompt excerpt

No prompt excerpt was provided by the source.

Sponsored

CodeBuddy

Tencent Cloud AI coding assistant for IDE and CLI.

Try CodeBuddy ↗
Explore all →