Privacy policy
myspeakingbuddy is a voice-based English practice app made by MB “Jambo Grupė”, a small partnership registered in Lithuania (company code 303081220, registered office Sausio 13-osios g. 2, Vilnius, Lithuania) — the data controller for everything described here. This page explains, in plain language, what information we handle when you use the app, why, and the control you have over it. If anything is unclear, write to us: hello@myspeakingbuddy.com.
The short version
- We never store your voice audio. It is streamed, transcribed in real time, and discarded — and we explicitly opt every request out of our speech provider's data-retention and model-improvement programme, so they keep it only long enough to transcribe it.
- We keep text transcripts of your conversations and the “memories” Sage learns about you — so your buddy can remember you. You can see and delete every one of them, any time, inside the app.
- We never use your data to train AI models, and we configure each of our providers not to either — the details are in the processor list below.
- We never see your card details — payments are handled entirely by Stripe.
What we collect and why
| Data | Why | Kept |
|---|---|---|
| Account (email, name you choose) | Sign-in, addressing you by name | Until you delete your account |
| Voice audio | Live transcription only | Not stored — processed in real time, then gone |
| Conversation transcripts (text) | Session continuity, your reflections, word study | Until you delete them or your account |
| Learner memories (facts Sage notices) | Making conversations personal across sessions | Until you “forget” them in-app or delete your account |
| Learning data (streaks, vocabulary, progress) | Your progress screens and spaced repetition | Until you delete your account |
| Billing status (subscription state from Stripe) | Knowing your membership is active | While you have a membership record |
Who processes it for us
We run on a small set of specialised services, each receiving only what it needs to do its job:
- Cloudflare — hosting, database, and audio streaming infrastructure.
- Clerk — sign-in and account management (your email lives here).
- Deepgram — converts your speech to text in real time. We opt every single request out of Deepgram's data-retention and model-improvement programme, so your audio is processed and discarded, not kept or learned from.
- Anthropic — the AI (Claude) that powers Sage's side of the conversation. Data sent via the API is not used to train models.
- Cartesia — converts Sage's replies to speech. Cartesia receives only the text Sage says (never your voice), and we have asked Cartesia to exclude our content from model training.
- Stripe — payments. Your card number never touches our servers.
Where your data goes
Some of these providers process data in the United States. Where that happens, the transfer is covered by the EU–US Data Privacy Framework or by Standard Contractual Clauses — the safeguards EU law requires.
Your controls and rights
- See everything Sage knows on the “What Buddy knows” screen — and forget any memory with one tap.
- Delete your data or your whole account: email hello@myspeakingbuddy.com and we'll erase your records within 30 days. (In-app self-serve deletion is coming.) One exception we can't waive: invoices and billing records are kept for as long as Lithuanian accounting law requires.
- Export: on request, we'll send you a copy of your stored data in a portable format.
- Everything else GDPR gives you: you can ask us to correct, restrict, or object to any processing of your data — same address. And if you think we've handled your data badly, you can complain to the Lithuanian data protection authority (VDAI) or the authority in your own country.
Legal basis (EU/GDPR)
We process your data to perform our contract with you (running the service you signed up for), and on legitimate interest for basic service security and improvement. We don't sell your data, run third-party advertising, or use tracking cookies — the only cookies are those needed to keep you signed in.
Age
myspeakingbuddy is intended for learners aged 16 and over. If you are younger, please ask a parent, guardian, or teacher to set things up with us directly.
Changes
If this policy changes in a way that matters, we'll tell you in the app before it takes effect. The date at the top always reflects the current version.