1Who is responsible for your data
SpanVoice is operated by ARKNET PTY LTD, Brisbane, Queensland, Australia. For the account information we hold, we are the controller.
For what is said at your events, the roles are the other way round. You are the controller and we are the processor. It is your room, your speakers and your audience; we process what comes through on your instructions and according to the settings you choose. Where the law where you are requires people to be told that their speech is being translated, captioned or kept, telling them is yours to do.
2Nobody in your audience signs up, and we do not identify them
This is the most important section on the page, so it comes first.
Somebody in your room opens a link or scans a code and picks a language. They do not make an account. We do not ask for a name, an email address, a telephone number or a sign-in. We do not place advertising or tracking on the audience page and we do not build a profile of anybody who reads along.
Their phone remembers two things, in the phone’s own storage, not ours: the language they chose, so they do not have to pick it again if the page reloads during the event, and their display preferences such as text size. That memory lives in their browser, lasts a few hours, never reaches us, and disappears when they clear their browser data.
Our servers see the ordinary technical facts any web server sees in order to answer a request — an IP address, a browser type, times of connection. We use them to run and secure the service and we do not use them to identify individual members of an audience.
Two things deserve to be said plainly, because “no account” is not the same as “no information”. The first is those technical facts above. The second is that what a speaker says is turned into text and that text is kept for a while — so if somebody’s name is spoken aloud, their name is in the transcript, whether or not they ever opened the link. Neither is tied to anyone’s identity by us, because we hold no identities for an audience. But neither is nothing, and the next two sections say exactly what happens to each.
3We never record the audio
Audio arrives from the operator, is processed as it flows, and is gone. It is never saved to a disk, on our servers or anywhere else. There is no recording feature, no archive of what a room sounded like, and no way for us to play back an event, because nothing was kept to play.
Spoken translation sent to listeners is generated as it is needed and held briefly in memory so that a repeated line does not have to be made twice. It is not stored as files and it does not survive the service restarting.
The distinction matters on a product whose whole job is turning speech into text, so here it is without the wordplay: we keep no recording of your event, and we do keep the text produced from it. The next section is about that text.
If you need a recording, make it yourself with whatever you already use. That is deliberately not our job.
4Transcripts, and how long they stay
Text is different from audio, because an account needs to be able to look back at what was said and check what was charged.
For each line, we keep the venue it belonged to, which session it was part of, the time, the language spoken, the language it was turned into, the original text and the translated text. Nothing about who in the audience read it, because we do not know.
Your venue chooses how long transcripts are kept: three, six or twelve months. Three is the default for a new account. A job runs every night and permanently deletes everything older than the limit you have set. If you shorten the limit, the next night’s run removes anything already past the new one — shortening it is not just a setting for the future.
You can change the setting at any time in the admin panel, under your organisation. You can also export transcripts from there while they exist.
5Account information
For the people who hold and run an account, we keep what is needed to have an account at all: a name and email address for each person signed in, a password stored only as a cryptographic hash that cannot be read back, the organisation’s name and billing address, the venues set up under it with their settings and branding, and a record of sessions, usage and charges.
We use it to provide the service, to bill correctly, to tell you about things that affect your account, and to keep the service secure. We do not sell it. We do not send marketing to your audience, because we have no way to contact them.
We also keep an audit record of significant changes made inside an account — who changed a setting and when — because an account with more than one person in it needs to be able to answer that question.
6Payments
Card payments are handled by our payment provider, which is certified to the card industry’s security standard. The card details go to them, not to us. We never receive or store your full card number. What we keep is what we need to show you on an invoice: the last four digits, the card type, the expiry, and the record of what was charged and when.
7Where it runs, and where your data is
The service runs from Australia, Europe and the United States. A venue is served from one of those, so that audio does not have to cross the world and arrive late.
Your account and billing records are held in Australia. Transcripts are held in the region your venue is served from. If you have a requirement about where data must stay, tell us before you set a venue up, because a venue’s region is chosen when the venue is created and cannot be changed afterwards.
Where data moves between these regions, it is because we are operating one service across them, and it is protected by the arrangements described in the next section.
8Who else processes it
We use a small number of suppliers, each under a contract that requires them to process data only on our instructions and to protect it:
- a speech and translation provider, which receives the audio of a session and returns text, and which receives the text and returns spoken audio;
- a payment provider, which handles card details and charges;
- hosting providers in each region, which run the servers;
- an email provider, for the messages we send to account holders.
They are processors, not customers. None of them is permitted to use what passes through for their own purposes, and none receives your transcripts as an archive.
We do not give your data to anyone else unless you ask us to, or unless we are legally required to, in which case we will tell you unless we are forbidden from doing so.
9Nothing is used to train anything
What is said at your events is not used to train or evaluate any model, by us or by the suppliers we use. The vocabulary you give a venue — names, places, your own terms — is used for that venue and no other.
One honest exception, because it is the kind of thing a policy should not hide in silence: if you choose to send us an example in a support message, we read it in order to fix what you reported. That is content you sent us deliberately, it is not taken from your events, and we keep it only as long as it takes to resolve the thing you told us about.
11Your rights
Depending on where you live, you may have the right to ask for a copy of what we hold about you, to have it corrected, to have it deleted, to object to or restrict what we do with it, to receive it in a portable form, and to complain to a regulator.
If you hold an account, most of this you can do yourself in the admin panel, which is faster than asking us: see what is held, export it, change it, shorten retention, or close the account. For anything the panel does not cover, use the contact form and we will respond within the time the law requires.
If you attended an event and want to know what was kept about you, the venue that ran the event is the right place to ask, not us. They hold the relationship with you and they control the settings; we act on their instructions. Send it to them and we will help them answer it.
In Australia you can complain to the Office of the Australian Information Commissioner. In Europe and the United Kingdom you can complain to your national supervisory authority. We would rather you came to us first so we can fix it.
12Keeping it safe
Traffic between your room, our servers and your audience is encrypted in transit. Passwords are stored hashed and cannot be read back by us. Access to production systems is limited to the people who need it.
No service can promise it will never be breached. If one happens and it affects you, we will tell you and the relevant regulator as the law requires, and we will tell you what we know rather than waiting until we know everything.
13Children
Accounts are held by organisations and run by adults. Children do not sign up for anything, and the audience page asks nobody for a name, an address or an age.
That is not the whole answer, and we would rather give you the whole answer. Events at schools are an ordinary use of this service, and when a student or a parent speaks, what they say becomes text in the transcript exactly as anyone else’s words do. A child does not have to open the link, or own a phone, to end up in a record of what was said in the room.
In that setting the school decides whether to run the service at all, how long transcripts are kept, and what the people in the room are told. We act on the school’s instructions. The school’s own policy governs its relationship with its students; it does not replace this page, which describes what we do.
If you are a parent or a student and want to know what was kept, ask the school first — it holds both the relationship and the setting, and we will help it answer. If your question is about something we decide rather than something the school decides, come to us directly and we will not send you away.
14Changes to this policy
We may change this policy. The date at the top says when it last changed. If a change materially affects what we do with your data, we will tell the account by email before it takes effect.
15Contact
For any privacy question, or to exercise a right, use the contact form and say that it is a privacy request so it reaches the right person.
SpanVoice is operated by ARKNET PTY LTD, Brisbane, Queensland, Australia.