How to batch-produce voiceovers from a watched folder with VoiceStudio

Who it is forContent team producing weekly narration for dozens of short scripts

The problemRendering each script by hand in a GUI is slow and repetitive.

The outcomeScripts dropped into a folder come out as finished audio using a saved recipe.

What you need first

• macOS, Windows or Linux desktop (Intel Mac runs the UI only and needs a remote backend)
• NVIDIA GPU (CUDA) or Apple Silicon (Metal) recommended; CPU-only works but is slower and needs about 5 GB free disk
• Git, Node.js 22+, Bun and Rust/Cargo only if building from source
• Speech models are downloaded on demand through the app; no paid keys or accounts required

Steps

1. Install the app and set up one voice and settings you like.
2. Save that setup as a reusable recipe, as described in the repo docs.
3. Point a watch folder at your scripts directory.
4. Drop in a few scripts and check that each produces an audio file.

Prompt for your agent

Using debpalash/VoiceStudio, configure a recipe and watch folder for <scripts directory> so each new script renders to audio. Verify one test script produces output with the expected voice.

Caveat. The README names recipes and watch folders but gives no setup commands here, so follow the in-repo docs and expect to experiment.

• Markdown version
• JSON version