Privacy Policy — Poducasto

Last updated: November 4, 2026

No account, no personal data collection

Poducasto transcribes audio and video files with AI, entirely on your device. It does not require an account, and it does not collect, sell, or share your personal information.

Your files and transcripts stay on your device

You choose which file to import using Android's own file picker, so the app only gets access to the files you pick. The audio is decoded and transcribed locally by an on-device speech recognition model (Whisper, running through whisper.cpp). The resulting transcripts, and a local playback copy of the audio when needed, are stored only in the app's private storage on your device — nothing is uploaded to any server. Deleting a transcript in the app removes it, along with that local copy; your original file is never modified or deleted.

Downloads and internet use

The first time you transcribe, Poducasto downloads the Whisper speech recognition model file from Hugging Face (huggingface.co) over a secure (HTTPS) connection. It's a plain file download: no audio, text, or personal data is sent. If you use subtitle translation, the app downloads an on-device translation language model from Google (ML Kit) — the text itself is translated locally on your device and is never sent anywhere. Like all ML Kit apps, the ML Kit library may send Google anonymous technical information (device and app details, API usage and performance metrics) to maintain and improve the service; this never includes your audio, transcripts or personal data. If you choose an AI voice to create podcasts from text, the app downloads that voice model from GitHub (the sherpa-onnx project, github.com), also as a plain file download; the speech is then generated on your device. Podcast packs you buy are downloaded through Google Play. Apart from these downloads and purchases, the app works fully offline.

Purchases

Podcast packs are optional one-time purchases handled entirely by Google Play. Poducasto never sees or stores your payment details: it only receives from Google Play whether you own each pack, so it can unlock and download it. Google processes the payment under the Google Privacy Policy and the Google Play Terms of Service. Once added, the pack episodes are stored in your library on your device like any other podcast.

Permissions used

Poducasto uses the following device permissions: internet access (only for the downloads and purchases described above), Google Play billing (to buy podcast packs), notifications (to show transcription progress), foreground services and wake lock (so a long transcription keeps running while the screen is off or you use other apps, and so playback keeps going with the system media controls). The battery button in Settings only opens Android's own battery optimization screen; it doesn't use any extra permission. No analytics, advertising, or tracking services are used.

Legal notice

Poducasto is a tool for transcribing files you already have on your device. Apart from the sample podcasts and the podcast packs, which are created by Mendelatech with AI-generated voices and licensed for your personal use, it doesn't host, store, or distribute any content. You are responsible for making sure you have the right to use the files you transcribe; transcripts are meant for personal use.

Contact

Questions or requests regarding your data? Contact: mendelatech.greeting513@passinbox.com

Open source licenses

Our links