How Clip Vuno processes files in your browser

Opening a file in Clip Vuno does not start a media upload. Available tools read it into browser memory, process it on your device, and create a downloadable result.

Three ways a file is processed

Image resizing, cropping and format conversion use browser image and canvas APIs. Audio and video tools use FFmpeg compiled to WebAssembly. The first media operation downloads an engine of about 31 MB from Clip Vuno, then runs it in your browser.

Local AI tools download a task-specific model when you start a task. The current speech workflows use Whisper, portrait background removal uses MODNet, and local English voice generation uses SpeechT5. Those model downloads are separate from your media: your file is not sent to an AI inference server.

What stays on your device

Your input media, intermediate processing data, AI transcript and generated output remain in the browser for the available local workflows. A download link points to a file generated in browser memory. Save the output before closing the tab.

What the website does send

Clip Vuno still makes network requests to load the website, media engine and AI models. AI runtime and model assets may come from jsDelivr and Hugging Face. The active-visitor counter sends a random browser ID and activity time. The files-processed counter sends an anonymous event ID after a successful output. Neither counter sends media content or filenames.

Why large files can fail

WebAssembly processing needs memory for the source, decoded frames, audio buffers and output. A compressed input may expand significantly while processing. The media tools accept inputs up to 150 MB, but a smaller file can still exceed a device's available memory. Try a shorter segment, lower-resolution source, or a desktop browser.

Internet and offline use