Vociamo privacy notice
Version of 28 October 2026 (first version: 10 September 2026; changed on 1 October: where the OVHcloud server is, and which emails we send; on 16 October: the address to write to, now privacy@vociamo.app, and the transcript, translations included, readable by those who take part and downloadable if the organisation allows it; on 19 October: the notice covers every kind of event, tours, lectures and conferences, and speaks of guest speakers without an account and of staff names shown to the audience; on 28 October: Gemini, the glossary helper that is switched off today, has a line of its own, because it is not in Google's European regions). It has not been reviewed by a lawyer (owner's decision of 10 September: published as is for now, to be reviewed when one is available). The technical facts (which data, where, for how long) are taken from Vociamo's code and database and are true on the date above; the legal parts are written with the care of someone who is not a lawyer.
1. Who processes your data
Data controller: Alessandro Nardoni, via XXV Aprile 33, 50065 Pontassieve (FI), Italy, email privacy@vociamo.app.
This is the heading chosen on 10 September 2026 for the trial events. Before the first paid event the controller will become a sole trader or a company and this line will change.
2. What Vociamo is, in one sentence
Vociamo is a service for live events, that is a tour, a lecture or a conference: whoever speaks uses their own language, and every participant follows on their own phone, through an earphone or by reading captions, a machine translation in the language they chose. Nothing to install: you scan a QR code and a web page opens.
We process data about three kinds of people: operators (the organisation using Vociamo and its staff, those who speak and those who moderate, who have an account), guest speakers (people who speak at an event without an account, through a personal link) and the audience (the people taking part in the event, who have no account).
3. Which data, why, for how long
3a. Operators (with an account)
| Data | Why | For how long |
|---|---|---|
| Email, password (stored only as a hash, never in clear), display name | to let you in and recognise you | as long as the account exists |
| Organisation and role (owner, admin, member; speaker or moderator of an event) | to decide what you may see and do | as long as the account exists |
| The events you create: title, kind, languages, sections (stops, topics or talks) with their texts, glossary terms with approved translations, the guest speakers you add | it is the service itself | until you delete them or the account |
| Backup recordings of sections, if you choose to upload them: audio prepared in advance, which may contain your voice, to play when live translation is not enough (there are none by default) | to have a standby voice in each language | until you delete them |
| The recognised text of what you say during the event, sentence by sentence, and its translation into each language of the event | to translate it and show captions; for those taking part to read the whole text while the event is running and, if your organisation allows it, to be downloaded by the phones that took part, also after the end; then for diagnosis if something went wrong | 30 days from when each sentence was said, then deleted automatically every night |
| Usage counters per event (minutes spoken, characters translated, characters synthesised), without prices | for the usage report on the "Usage" page and for invoicing | kept even after the event is deleted |
| The action log (who opened an event, who added whom, who closed it) | for security and to reconstruct what happened | 12 months, then deleted automatically every night |
| The emails we send you (address confirmation) | to make the account work | the email provider keeps a sending log for a limited time |
Your display name is shown to the audience when you speak at an event where more than one person speaks: it appears above your sentences in the captions and in the text the audience can read again or download. If you have not written a display name, the word of the kind of event ("Guide", "Speaker") appears instead. Your email is never shown to the audience. The display name is changed in Settings › Account.
The audience's transcript download is off until your organisation turns it on for an event; only the organisation's owner and administrators can turn it on. When it is on, those who took part can save the text on their own phone: that copy is theirs, and the deletion after 30 days does not reach it.
Your voice is not recorded during the event. Backup recordings are files you upload beforehand, not recordings of the event. During the event the audio passes through our server and through Google's speech recognition, translation and speech synthesis services, which process it on the fly and do not keep it.
3b. Guest speakers (no account)
A guest speaker is a person who speaks at an event without having an account: the organiser adds them with a name and a colour, and sends them a personal link that works for that event only.
| Data | Why | For how long |
|---|---|---|
| The name chosen by the organiser, and the colour | to show it to the audience above their sentences, and to the organiser in the console | for as long as the event exists: today no automatic deletion reaches it earlier |
| The personal link (stored in the database only as a hash), who created it, when, and whether and when it was revoked | to let them in and to be able to shut them out | as long as the event exists; the link stops working when the event ends or when it is revoked |
| The recognised text of what they say while they hold the floor, with its translations | as for operators: captions, reading again, download if allowed | 30 days from when each sentence was said |
A guest speaker's voice is listened to only while they hold the floor: their microphone left open while somebody else speaks is not transcribed. As for operators, the audio is not recorded.
3c. The audience (no account)
| Data | Why | For how long |
|---|---|---|
| A random key generated by your phone the first time you join (stored in the database only as a hash) | to recognise the same phone if it reloads the page, without knowing who you are; and, after the event, to recognise that the phone was there if the organisation allows the text to be downloaded | see below |
| The language you chose and your preference between voice and captions | to give you the right channel | see below |
| When you joined and left the event | to count how many people listen to each language | see below |
| The questions you write to the speakers and your "raised hands" | to deliver them to the speakers, translated into their language | 90 days, then deleted automatically every night |
| Your phone's IP address, in the technical logs of the providers that host the service | to make the connection work and for security | a limited time set by the providers (see section 5) |
The "see below" rows (event participation) are kept for 90 days, like the questions, then deleted automatically every night.
We do not ask for your name, email or phone number. We do not know who you are: we know that a phone listened in German from 10:12 to 11:40 and wrote two questions.
Your voice never enters Vociamo: the audience only listens. Questions to the speakers are written.
3d. Cookies and browser storage
Technical cookies and entries only; no advertising tracking, no third-party analytics:
- for operators: the account session cookie (provided by Supabase),
lip_locale(interface language),lip_org(last organisation opened),lip_theme(light or dark); - for the audience: in the phone's storage the anonymous key
lip-anonymous-keyand the listening preferences (lip-audience-*), which stay on the phone.
4. Legal basis
- Operators: the service contract with your organisation (Article 6(1)(b) of Regulation (EU) 2016/679, the "GDPR").
- Guest speakers: the event organiser's legitimate interest in letting their guests speak and in showing the audience who is speaking (Article 6(1)(f)); the guest knows their voice is transcribed and translated, because their page says so, and can stop speaking at any time.
- The audience: the event organiser's legitimate interest in being understood by the participants and ours in running the service (Article 6(1)(f)); a participant leaves no identifying data and can stop listening at any time by closing the page.
- Backup recordings, which may contain a speaker's voice: the explicit choice of whoever uploads them (Article 6(1)(a)).
5. Who sees the data besides us (providers)
Vociamo runs on other companies' services, which process data on our behalf:
| Provider | What it does | Where |
|---|---|---|
| Supabase | the database and the accounts | Frankfurt, Germany |
| Vercel | hosts the web app | app executed in Frankfurt; worldwide delivery network for static pages |
| OVHcloud | the server that processes the audio and the switchboard that distributes it to phones | Frankfurt, Germany |
| Google Cloud | speech recognition, translation and speech synthesis; the glossary's automatic proposals made with translation | processing in Google's European regions (eu, europe-west1); Google's standard contractual clauses cover the cases where Google processes data outside the Union |
| Google Gemini | automated help for the glossary, switched off today; it would receive only the glossary's terms, never the voice | if it is ever switched on, processing is not guaranteed to stay in Europe: the service we would use has no European region; Google's standard contractual clauses cover transfers outside the Union |
| Resend | sending emails (confirmation) | Ireland |
| OpenAI | only if an organisation asks for "direct" translation with that provider, off today | United States, with standard contractual clauses |
No data is sold or passed to third parties for advertising.
6. Your rights
You can ask us at any time: what data we hold about you (access), to correct it, to delete it, to restrict its use, to receive it in a readable format (portability), to object to processing based on legitimate interest. An email to privacy@vociamo.app is enough; we answer within 30 days. You can also lodge a complaint with the Italian supervisory authority, the Garante per la protezione dei dati personali (www.garanteprivacy.it).
For a participant, who leaves no identifying data, deletion works by naming the event and the time: we delete the participation and the questions written in that window.
A guest speaker can ask at the same address, naming the event, for their name and their sentences to be deleted.
An operator who wants to delete their account, or an organisation that wants to close, asks at the same address: there is no button in the product to do it alone, and the request is carried out within 30 days.
7. Minors
Vociamo is not aimed at minors. A minor taking part in an event does so under the responsibility of the organiser and of the person accompanying them; we collect no data that would identify them.
8. Security, in short
Passwords are stored only as hashes; signing in requires email confirmation; every piece of data is protected by database rules that stop one organisation from seeing another's; the audience enters with an event code, and each guest speaker with a personal link, which the organiser can revoke at any time; connections are encrypted (HTTPS, WSS). The providers in section 5 are all subject to the GDPR or bound by standard contractual clauses.
9. Changes
If we change something substantial (a new provider, a different retention period), we update this page and notify operators by email. The date at the top says when the last version was written.