This policy covers one feature of the ViralDNA.io platform: ViralDNA Text to Speech, the tool at viraldna.io/text-to-speech that turns English text into a voiceover you can download as a WAV or MP3 file.
It has its own policy because your text never leaves your device. The voice is generated by a model that runs inside your browser. Our server never receives your text or the audio made from it.
Everything else about your ViralDNA account, including sign-in, billing and the other tools, is covered by the Platform Privacy Policy. This document adds to it and does not replace it.
The short version.
You paste text, your browser turns it into speech on your own computer, and the WAV or MP3 file is saved straight to your device. The voice model and the voice you pick are downloaded from Hugging Face. Your last 10 voiceovers are kept in your own browser so you can play them again. We store nothing about what you wrote or generated.
The voice samples on the page are short recordings served by our website. Playing one sends nothing about you beyond an ordinary file request.
When you press the button, your browser runs Phonon 31M, an open text-to-speech model published by Gradium, in a background worker. The first time, it downloads the model files (about 65 MB) and the file for the voice you chose (about 256 KB) from Hugging Face and keeps them in its own cache, so later runs do not download them again. The speech is then generated on your computer, sentence by sentence, and assembled into a WAV file in your browser's memory. If you choose MP3, the WAV is converted to MP3 in your browser as well, by the LAME encoder running in a background worker; nothing is uploaded for it.
To decide how many characters you can convert, the page asks our server whether you are signed in and whether your plan is Pro. That check reads your session and your plan; it receives no text and writes nothing.
On our side, nothing about your text or audio. When a voiceover finishes, the page sends our own first-party, cookieless usage statistics one event saying that the tool was used, with the page address and the page language, as described in section 2.6 of the Platform Privacy Policy. It contains no text, no audio, no voice choice and no length.
On your device:
viraldna-tts): the audio, the voice, the first sentence of the text, the length and the time it was made. Each entry is tagged with your account ID, so another account signed in on the same browser does not see it. When an 11th voiceover is made, the oldest one is deleted. You can delete any entry from the list on the page. These entries are never sent to us and do not appear on your other devices.viraldna_tts_free_), together with the date. It holds a number, never your text.viraldna_tts_prefs in your browser's local storage, so they are selected next time.Clearing this site's data in your browser removes all of the above.
As with any website, our hosting provider records that pages and files were requested (time, address, status, IP address and browser user agent) for a short period, for security and troubleshooting.
The voices are designed voices, not copies of real people. Please do not use generated speech to impersonate a real person or to mislead listeners about who is speaking, and tell your audience that a voice is synthetic where that matters to them.
If we ever change how this feature handles data, we will update this page and change the date at the top before the change takes effect. If a change ever meant that your text would be transmitted or stored, we would say so in plain language on the feature page itself, not only here.
Questions about this policy: support@viraldna.io.