TTS, or injecting custom content

[Skyb0rg007]

One feature of Pocket is text-to-speech: every text article can be read aloud to you by an AI. There are plenty of self-hosted TTS cli apps, but there is an issue integrating them with Miniflux. It would be really nice to write a custom script that generates an audio file for every item in a feed, and use custom HTML+CSS to add the content to the screen, but this doesn’t seem to be available. A custom button would allow a similar flow.

Is this something possible to add to my Miniflux instance today? Or would this require a code addition as I suspect?