SameVoice exists so families can hear each other across languages, each person in their own voice. This version (0.2.0) enrolls your voice on your device, and — if you choose to create a free account — connects you with family by invite code for messaging and app-to-app calls, with voice messages translated by our own engine. The short version: your enrollment never leaves your phone; what the account features must hold is listed plainly below; nothing is sold, there are no ads and no trackers, and calls are never recorded.
01 What we collect, and why
Used without an account, SameVoice collects nothing: enrollment, your people and on-device voice messages never leave the phone, and there is no analytics, tracking, or advertising software in the app — ours or anyone else's — and no advertising or device identifiers.
If you create a SameVoice account (the same account system as vidnexa.com), we hold only what the account features need to work:
- Account details — your username, email address, date of birth, and a hashed password (never the password itself), so you can sign in and recover the account.
- Connections — who you connected with, by invite code. There is no directory and no search: nobody can find you unless you share your code.
- Messages you send in the app — text and voice messages are stored on VidNexa's server so they can be delivered when the other person's phone next opens SameVoice, and so voice messages can be translated (section 05). Only the two participants can read or play them.
- Call records — who called whom, when, and for how long, so the call history is real. Never the call's content (section 07).
- Presence — while your app is open and signed in, contacts you accepted can see that you are online.
Nothing in SameVoice is sold or shared with data brokers, advertisers, or analytics companies. The app also briefly contacts VidNexa's own server to download its configuration (which optional features are available, and any announcement): that request carries the app's own key and, when you are signed in, your account token — nothing else. Voice enrollment, your people and on-device messages work fully offline and are never blocked by that download.
02 On the device only
Your voice enrollment recordings and the voice data made from them, your people (their names and language pairs), and the voice messages you keep on the phone are stored only in the app's private storage on that device. Enrollment audio is never uploaded to VidNexa or to anyone else.
- Your enrolled voice, any person, and any on-device voice message can be deleted inside the app at any time; deleting removes the recordings and their details together.
- Removing the app removes all of its on-device data with it.
- A voice message leaves the phone only when you send it: inside SameVoice to a connected person (section 05), or by the share sheet, using your own tools.
03 The microphone & camera
The app asks for microphone permission for: enrolling your voice, recording a voice message, and speaking on a call. If you turn captions on during a call, short audio chunks (about five seconds) are sent to our translation engine and are not stored after they are captioned. The camera is used only during a video call you start or answer — never at any other time. Both are used only after you tap to start. If a permission is refused, the rest of the app still works; only the matching feature is unavailable.
04 Your voice and consent
Your voice is personal — it can only ever be yours. Before enrollment, SameVoice shows a plain-language explanation: that you will record your own voice, live; that your voice data stays on this device; and that your voice may be used to speak your translated words, only under your consent. Enrollment cannot begin until you agree.
- Your name, your agreement, and the time of that agreement are stored on the device alongside your enrollment.
- Consent can be withdrawn at any time in the app. Withdrawal is enforced, not just recorded: the translation engine refuses a withdrawn voice outright — it is never synthesized, for anyone.
- Nobody can enroll a voice for someone else — the person whose voice it is must record it, live, themselves.
- SameVoice will never sell or license your voice, and no voice is ever used to train general-purpose AI models.
05 People, messages & translation
The people you keep on this phone — a person's name and the languages you speak with them — are stored in small files on the device and are never transmitted. When you connect with someone, the connection itself (both accounts, and who accepted) is held on VidNexa's server, as section 01 describes.
Messages you send inside SameVoice are stored on VidNexa's server for delivery. A voice message is translated by VidNexa's own translation engine running on VidNexa-controlled infrastructure: the engine transcribes it, translates it, and — where your voice is enrolled with the engine under your consent — speaks the translation in your voice. The original is always kept and always playable, beside its translation. Only the sender and the recipient can fetch a message or its audio. To have your account and its server-held data deleted, contact us (section 08).
06 Who SameVoice is for
SameVoice is made for adults who want to talk with family across languages. It is not designed for children, it contains no advertising, and in this version it contains no purchases. The person enrolling a voice (section 04) is always the person whose voice it is.
07 Calls in this version
This version makes app-to-app voice and video calls between connected SameVoice users. It does not connect to any phone network — there is no calling of ordinary phone numbers in this version.
- Call audio and video travel directly between the two phones (WebRTC). When a direct path is impossible, the encrypted stream passes through VidNexa's own relay server, which cannot listen to it.
- Calls are never recorded — not by the app, not by the server. The only call data kept is the record in section 01 (who, when, how long).
- Call invitations and connection details pass through VidNexa's server, because the phones need a way to find each other.
- Live captions, when you turn them on, work as section 03 describes: short audio chunks to the engine, captioned and translated, not stored.
- A voice is used by the engine only while its owner's consent stands — withdrawn consent means the engine refuses that voice, always.
If we change this policy, we will update the date at the top of this page.
08 Contact us
Questions about SameVoice and privacy? We want to hear from you.
Privacy team
Reach us at privacy@vidnexa.com — we read every message and respond as soon as we can.
contact_support Contact usFor how the main VidNexa service handles data, see our Privacy Policy.