Privacy Policy
What A+ Farsi collects, why, and what stays on your child's device.
The short version
- No accounts, no sign-up — we never ask for an email, phone number, or password.
- No ads, no cross-app tracking, no advertising identifiers — nothing here follows you elsewhere.
- No third-party analytics or advertising SDKs of any kind — the only usage tracking is first-party code we wrote, sending anonymous events straight to our own server.
- A child's name, avatar, and progress stay on the device — never sent to us.
- Voice recordings are sent to a third-party AI service (OpenAI) only after a parent gives permission in the app, solely to score pronunciation and generate spoken examples.
- Anonymous usage statistics are off unless a parent turns them on in the app's "For parents" settings.
What we collect, and why
Voice recordings — only while practicing pronunciation, and only with a parent's permission. After a parent has allowed it in the app (see the third-party AI section below), when a child taps the microphone to practice speaking Farsi, the app records a short clip and sends it to our server, which forwards it to our speech provider, OpenAI, to transcribe it (so we can score the pronunciation) and to generate spoken word examples. We do not store these recordings to build a profile. Under our provider's API terms, audio is not used to train its models and is retained only briefly, if at all, for abuse monitoring, then deleted. Audio is tied to a request, not to a child's identity.
Anonymous usage statistics (off by default). Only if a parent turns this on under Me → For parents, the app sends aggregate, anonymous events — for example "app opened" or "viewed the Alphabet screen." These are keyed by a random identifier created on the device — not your name, not your Apple ID, and not a device or advertising ID. We also receive the platform and OS version. This data is not linked to any identity and not used to track you.
Information stored only on the device (not collected by us). The profile name and avatar you choose, your learning progress, and your language setting are saved locally and never transmitted off the device. Deleting a profile or the app removes all of it.
Illustrations. Most word illustrations are bundled inside the app itself and load with no network request. A smaller set of everyday-object photos are requested from Wikimedia Commons at view time — as with any web image, that reveals your device's IP address to that host. We never send Wikimedia any personal information alongside the request.
Third-party AI service (OpenAI) — what is sent, to whom, and with what permission
What data is sent. When a child taps the speak button on the Practice screen, the app records a short audio clip (a few seconds) of the child saying the word. That clip is sent together with the text of the word being practiced. When the app plays a word that has no pre-recorded audio (for example, a word the child typed into "My words"), the text of that word is sent to generate speech. No name, nickname, avatar, photo, contact information, device identifier, or advertising identifier is ever sent.
How it is collected. Only through the device microphone, only while the child is actively recording on the Practice screen, and only after the iOS microphone permission has been granted.
Who it is sent to. The data travels over an encrypted connection to our own server (hosted on Google Cloud), which forwards it to OpenAI, L.L.C., a third-party AI service, through OpenAI's API.
All uses of this data. (1) Converting the child's speech to text so the app can compare it with the target word and show a pronunciation score; (2) generating spoken audio of words. The data is not used for advertising, profiling, tracking, or any other purpose.
Permission first. Before anything is sent, the app shows a screen explaining what is sent, who it is sent to, and why, and asks for permission. Because the app is for children, permission can only be given by a parent after passing a parental gate. If permission is not given, nothing is sent: the rest of the app works normally and only speaking practice is turned off. A parent can change this choice at any time under Me → For parents.
Equal protection. OpenAI processes this data under its API data-usage commitments and data processing terms, which provide the same or equal protection as this policy: API data is not used to train OpenAI's models, is used only to provide the requested service, and is retained only temporarily (up to 30 days) for abuse monitoring before deletion. We do not store recordings on our server.
Third-party analytics & advertising
Third-party analytics: none. A+ Farsi does not integrate any third-party analytics SDK — no Firebase, Google Analytics, Amplitude, Mixpanel, or similar. The "anonymous usage statistics" described above are first-party only: code we wrote ourselves, sending events directly to a server we operate. No analytics vendor ever receives this data.
Third-party advertising: none. The app contains no advertising SDK, shows no ads, and does not participate in any ad network, ad exchange, or ad attribution service.
Data shared with third parties. The only outside party that ever receives data from the app is OpenAI, and only for one purpose: transcribing a child's pronunciation-practice recording and generating spoken Farsi audio. This is a functional service call, not analytics or advertising, and OpenAI does not use this audio to train its models. Google Cloud separately hosts our server and anonymous-analytics storage as an infrastructure processor. Neither party receives data for advertising, profiling, or any purpose beyond running the app's features.
Data collected beyond the above: none. We do not collect device identifiers, advertising IDs, contacts, location, browsing history, or any data beyond the voice recordings and anonymous usage events described in this policy.
How this maps to Apple's privacy labels
| Data type | Purpose | Linked to you | Used to track you |
|---|---|---|---|
| Audio data | App functionality | No | No |
| Product interaction | Analytics | No | No |
| Random device identifier | Analytics | No | No |
Children's privacy
A+ Farsi is built for children and families. We do not knowingly collect personal information from children beyond the anonymous, functional data described above, and we never condition play on providing personal information. If you're a parent or guardian and believe we've collected personal information from your child that we shouldn't have, contact us at judysu1983@icloud.com and we'll delete it.
Sharing, retention & security
We do not sell or rent data. We share data only with the infrastructure providers that operate the app on our behalf — Google Cloud (hosts our server and anonymous-analytics storage) and OpenAI (speech transcription and synthesis) — acting as processors for those functions only.
Anonymous usage events are retained for up to 14 months, then deleted or aggregated. We do not retain voice recordings; our speech provider's brief retention is described above. Requests are sent over encrypted connections (HTTPS/TLS), and our API key is held only on our server — never inside the app.
Your choices & changes
Because we don't hold an account for you, the on-device data (name, avatar, progress) is fully under your control — delete the app to remove it. See Privacy Choices for more on data rights. For questions about anonymous analytics, or to request deletion of data tied to your device's random identifier, contact judysu1983@icloud.com. We may update this policy; we'll change the effective date above and note material changes in the app or store listing.