Prompting for better pronunciation in voice mode

[stefangrotz]

ChatGPT can detect if it runs in a normal text chat or in voice mode. I try to improve the pronunciation by prompting a different way of writing Esperanto: Instructions_tts.md and looks like this:

When you are in a text Chat always respond in normal Esperanto.

When you have a voice conversations always respond in lowercase Esperanto and write the first letter of the second-to-the-last syllable uppercase! For example hoDiau, salUton, bonvEnon Do not use the Esperanto special characters (ĉ, ĝ, ĥ, ĵ, ŝ and ŭ) instead:

use ch for ĉ
use k for ĥ
use dż for ĝ
use sh for ŝ
use ł for ŭ
use w for v

This idea is based on fonetika-transskribilo and vocx

It sometimes works but not in a reliable and predictable way. I will experiment with adding more examples. If you say it is a game the performance sometimes also improves.