NukoChineseShadowing ("the app") is an Android app for Taiwanese Mandarin shadowing
practice, developed by Nuko ("we", "our"). This page explains what data the app does —
and does not — handle.
Short version
The app's core features — automatic Zhuyin/Pinyin annotation, text-to-speech playback,
recording and comparing your own pronunciation, and managing your text list — all run
entirely on your device and send nothing anywhere. Data only ever leaves your device
through two optional features, and only if you explicitly turn them on:
AI pronunciation scoring: if you enter your own Gemini API key in Settings, your recorded audio (together with the target sentence text) is sent directly from your device to Google's Gemini API, using your own key, and a pronunciation/tone assessment is sent back. We do not operate any server, and we never receive or store this audio or text ourselves.
Camera text scanning (experimental): recognizing Chinese text from a photo uses Google's ML Kit on-device text recognition. This runs entirely on your device, and the image is never uploaded anywhere.
What's stored on your device
Your registered texts, bookmarks, recorded audio, and settings (including any Gemini API
key you enter) are stored only in the app's local storage on your device. None of this
leaves your device unless you explicitly use the "back up" feature to export a backup
file, which is written only to a location you choose, through Android's standard file
picker (the Storage Access Framework).
Permissions
Microphone (RECORD_AUDIO): used only when you tap the record button during shadowing practice, so you can compare your recording with the model audio, and — only if AI scoring is enabled — send it to the Gemini API as described above.
Internet (INTERNET): used only for the optional AI pronunciation scoring feature above. Every other feature works fully offline.
Camera: the app itself does not request camera permission. The camera text-scanning feature launches your device's own camera app to take the photo, and only receives the resulting image back.
Data collection and analytics
We do not collect:
Personal information (name, email, location, etc.)
Usage analytics or crash reports sent to any server
Advertising identifiers (the app contains no ads)
Any data at all, from anyone, including children (other than the Gemini API / ML Kit interactions described above)
Third-party services
The app integrates with exactly two third-party services, both provided by Google and
each governed by its own privacy policy:
Google Gemini API (only when AI scoring is enabled, using your own API key, communicating directly from your device)
Google ML Kit (text recognition, fully on-device with no network transmission)
No other third-party SDK, analytics platform, or ad network is included.
Children's privacy
Other than through the AI pronunciation scoring feature when you've enabled it with your
own API key, the app collects no data from anyone. It therefore does not knowingly
collect any information from children under 13 (or the relevant age in your country).
Changes to this policy
If this policy ever changes, this page will be updated and the "Last updated" date above
will change accordingly. Any such change would also be reflected in the app's in-app
description before release.
Contact
Questions about this policy or the app can be sent to: