Frequently asked questions
Updated September 15, 2026
Quick answers to the questions people ask most about AI Text to Speech, covering costs, OpenAI, bulk generation, privacy, and what happens to your files and data.
How much does it cost to generate audio?
OpenAI charges your account per character converted. The plugin estimates $0.012 per 1,000 characters for GPT-4o mini TTS and $0.015 for TTS-1, so a 1,000-word article (about 6,000 characters) costs roughly 7 to 9 US cents. Prices are set by OpenAI and can change, so check your OpenAI usage dashboard for real charges. Visitors listening to the audio costs nothing extra.
Do I need my own OpenAI account?
Yes. Both the free and PRO versions use your own OpenAI API key and bill usage to your OpenAI account. A ChatGPT subscription doesn’t include API credit; API billing is set up separately at platform.openai.com.
Can I generate audio for all my existing posts at once?
Not in version 3.2.0. Audio is generated per post from the editor, or automatically when a new post is published (PRO). For a back catalogue, open each post and click Generate TTS, starting with your most visited content.
Does the audio update when I edit a post?
No. Save your changes, then click Generate New in the AI TTS box.
Can I download the audio or use it elsewhere?
Yes. Files are standard MP3s. Use Copy File URL in the post editor, Download Audio File on the Generator page, or the Download link on the TTS Files page. Check OpenAI’s terms for how you can use generated audio.
Do I have to tell visitors the voice is AI?
OpenAI’s usage policies require a clear disclosure that the voice is AI-generated. The default player label does this. If you change or remove it, add your own disclosure near the player.
What data is sent to third parties?
- OpenAI receives the text being converted, the voice and style settings, when you generate audio.
- publish.twitter.com is contacted to fetch the text of embedded tweets during generation.
- Dropbox receives audio files if you use Dropbox storage (PRO).
- Google Charts loads in your browser on the Statistics admin page (PRO).
Nothing is sent to third parties when visitors play audio. Listening statistics store only counts and seconds, with no personal data. You may want to mention OpenAI and any other services you use in your privacy policy.
Are audio files private for draft posts?
No. Files are saved in the public uploads folder (or as public Dropbox links) with a name based on the post ID and title plus a short random string. Anyone who has the exact URL can play the file, even before the post is published. Don’t generate audio for confidential content.
What happens if I deactivate or delete the plugin?
The players disappear from your posts. Audio files in /wp-content/uploads/ai-text-to-speech/, your settings, and the audio links and statistics stored on posts are left in place, so reactivating restores everything. To remove it all, delete the audio from the TTS Files page before removing the plugin, then delete the ai-text-to-speech uploads folder.
Does it work with multisite?
Each site in a network has its own settings, API key and files. Activate and configure it per site.
Where do I get help?
Use Support in the plugin header bar. PRO users can contact support from inside WordPress; free users can post on the WordPress.org support forum. Include any message from the Error Log on the settings page.