Privacy
This page explains what happens to your information when you use ÒreAyò, in plain language. It covers the speech recognition preview on our homepage and the waitlist.
ÒreAyò is built by SOLAT Technologies Ltd, which is responsible for the information described here.
Last updated 30 August 2026
The short version
- • Your recording is sent to our servers, transcribed, and then discarded. We do not keep it.
- • We keep your recording only if you tick the box asking us to. That box is never ticked for you.
- • Showing a recording publicly needs a further, separate choice — and nothing is ever published without a person reviewing it first.
- • You do not need an account, and we do not ask for your name or email to try it.
- • We do not sell your information, and we do not use it for advertising.
Using the speech recognition preview
When you record or upload speech and press Transcribe, the audio is sent to our server, passed to our speech recognition service, converted to text, and returned to you.
The recording is not stored. It exists only for as long as it takes to produce the transcript, and is discarded once that is done. This happens automatically, whether or not you send us feedback.
We keep a small technical record of the request — which language was chosen, how long the audio was, how long processing took, and which version of the system handled it. This record does not contain your audio and is used to check the service is working.
If you tell us whether the transcript was right
After a transcription you can tell us whether it was accurate, and suggest a correction. This is optional.
If you do, we store:
- • the transcript the system produced
- • whether you said it was accurate
- • your correction, if you wrote one
- • the language, audio length, and system version
We use this to measure how well the system works with real speech, and to understand where it fails. Corrections are used for evaluation. They are not automatically added to the data used to train the system.
What happens to your recording
There are three separate situations, and each one needs its own choice by you.
1. By default — your recording is discarded
Your audio is used to produce the transcript and then thrown away. This is what happens unless you choose otherwise, and it needs no action from you.
2. If you tick the box — we keep it privately, to improve ÒreAyò
The recording is kept in private storage and used for research into speech recognition for Nigerian languages. It is not published, and it is not shared with anyone outside that work. Transcribing your speech never requires this.
3. Only with a further, separate choice — it may be shown publicly
We are building a small public gallery of real examples, so people can hear how ÒreAyò actually performs. A recording can only be considered for it if you explicitly opt in — separately from the choice above.
Opting in is not permission to publish. Every example is reviewed by a person first, examples are shown anonymously unless you have separately agreed to be named, and you can ask us to take one down at any time.
Please do not record information you would not want kept — such as passwords, bank details, or someone else's personal information — and only record other people with their agreement.
How long we keep things
- • Recordings you do not consent to: discarded immediately after transcription.
- • Feedback and corrections: kept while they remain useful for evaluating the system, and reviewed periodically.
- • Recordings you do consent to: kept until you ask us to delete them, or until they are no longer needed for research.
- • Published examples: shown until you ask us to take them down, or until we retire them.
- • Waitlist details: kept until you ask to be removed.
Asking us to delete something
You can ask us to delete feedback or a recording you submitted. Because the preview does not use accounts, we may need you to tell us roughly when you used it and in which language so we can find the right record.
These are two separate requests, and you can make either one:
- • Take it out of the public gallery. We remove the public example and its audio. Anything held privately for research stays as it is.
- • Delete it entirely. We remove the public example and the privately held recording and feedback.
Asking us to stop showing something publicly does not delete the private research copy unless you ask for that too — so please say which you want.
Email contact@solat.io and we will confirm once it is done.
The waitlist
If you join the waitlist, we collect the name, email address, and any note you provide, so we can send you progress updates. We use a third-party form service to receive these messages.
You can ask to be removed at any time by emailing contact@solat.io, and we will delete your details.
Others who handle information for us
We use a small number of established providers to run the site: cloud hosting for the website, a cloud provider for the speech recognition service, a database provider for feedback, and a product analytics service. They process information on our instructions.
Our analytics measures how the preview is used — such as how often transcription succeeds and which languages are chosen. It does not receive your audio, your transcript, or your corrections.
Children
The preview is intended for adults. If a child uses it with a parent or guardian, that adult is responsible for deciding whether to keep the recording.
Changes to this page
ÒreAyò is in active development, and what we collect may change as new capabilities appear. When it does, we will update this page and change the date at the top. We will not quietly start keeping recordings that people did not agree to.
Contact
If you have a question about privacy, want something deleted, or think we have got something wrong, please tell us — we would rather hear it. Email contact@solat.io.
Read more about our wider approach on the Safety & Privacy page, and the rules for using the preview in our Terms.