any plans for supporting screen readers?

[king-dahmanus]

hello there. Are there any plans to: First, speed up the models so they generate and synthesize speech in less than 20ms with a cpu, then making a synthdriver for the opensource, free and extensible nvda screen reader found at https://github.com/nvaccess/nvda so blind people can use any models available to read their texts? Thanks

[erogol]

We don't have any plans for now.