AI Text to Speech

How can we help?

Search the documentation or ask the AI agent anything about the plugin.

Free and PRO features compared

Updated September 15, 2026

The free version covers generating audio for posts, the Generator page, the audio player and the admin logs. PRO adds content controls, pronunciation tools, extra voice styles, automation, player styling, statistics and Dropbox storage. In the free version, PRO settings are visible but greyed out and marked (PRO).

Both versions use your own OpenAI API key, so OpenAI usage is billed to your OpenAI account either way.

FeatureFreePRO
Generate audio from the post editorYesYes
Generator page for any textYesYes
All 13 OpenAI voicesYesYes
Voice model choice (GPT-4o mini TTS or TTS-1)YesYes
Voice styles5 styles26 styles
Custom voice style instructionsNoYes
Automatic player on posts and [ai_tts_player] shortcodeYesYes
Include title, author and date in the audioYesYes
Upload your own audio file for a postYesYes
TTS Files page and API request logsYesYes
Post type and user role restrictionsYesYes
Global custom text before and after contentNoYes
Per-post content customisation (before/after or replace)NoYes
Custom pronunciationsNoYes
[tts say], [tts ignore] and [voice] shortcodesNoYes
Auto-generate audio when a post is publishedNoYes
Default voice and default voice styleNoYes
Player colours, height and default speedNoYes
Statistics and analyticsNoYes
Dropbox file storageNoYes

Free voice styles

Default, Conversational, Formal, Casual and Whispering are available without a licence. The other styles are listed in Voice styles and custom instructions.

If a PRO licence expires

Audio files you’ve already generated stay attached to their posts and keep playing. PRO settings become unavailable again, and the File Storage Location setting switches back to Local the next time the settings page loads, so new files are saved on your server. Files already stored in Dropbox keep their existing links.

Shortcodes such as [tts ignore] are still hidden from visitors on the front end, but they’re no longer processed during generation, so text inside them is read aloud if you regenerate.

Related

Was this article helpful?