Privacy Policy

This tool is designed with a strong commitment to user privacy.We do not collect any personal data.All audio processing for transcription happens either directly in your browser (desktop) or via a secure mobile flow that uses temporary Blob storage only for the time needed to complete transcription.

This tool is built with privacy as a core principle. How your data is handled depends on the device you are using:

• On Desktop: All transcription happens entirely inside your browser. No audio data, text, or any other personal information is ever sent to a server.
• On Mobile: Due to hardware limitations on phones, your audio is uploaded to a temporary Blob file, transcribed by Groq Cloud, and then deleted. Groq does not use your data to train their models.

Desktop Processing: When you upload audio on a desktop, it is decoded and processed entirely within your browser using the Whisper Small model running via WebAssembly or WebGPU. The AI model itself is downloaded from Hugging Face once and cached in your browser - after that, the tool works fully offline.

Mobile Processing: To prevent memory crashes on mobile browsers, your audio is chunked and uploaded directly from your device into a temporary Vercel Blob object. Depending on your Blob store mode, the server either hands Groq the Blob URL directly or streams it through a protected route. After transcription completes, the temporary upload is deleted and the text is returned to your device.

The Whisper model weights (approximately 150 MB) are cached in your browser's native cache after the first download. This is standard browser caching behavior (no different from a website caching its CSS or images). You can clear this cache at any time through your browser settings.

We only use a minimal localStorage flag to remember whether the model was loaded before, so the UI can skip a redundant first-load hint. We do not store audio, transcripts, personal data, or tracking identifiers in localStorage, sessionStorage, or IndexedDB.

Depending on your device, this application interacts with the following strict-privacy third parties:

• Hugging Face (Desktop): Used only to download the open-source Whisper model files on first use. Their privacy policy applies to this simple file download.
• Vercel Blob (Mobile Uploads): Used only as temporary storage so mobile audio can be uploaded directly. Temporary uploads are deleted after transcription completes.
• Groq Cloud (Mobile): Used as the processing engine for mobile transcriptions. Groq processes API requests ephemerally; they do not retain your audio data or use it to train their models. You can read the Groq Privacy Policy for more details.

This service does not knowingly collect any data from children or anyone else, as no data is collected at all.

If this policy ever changes, the updated version will be published at this URL with a revised "Last updated" date.

Questions? Reach out via LinkedIn.