I’m using espeak-ng to convert text to phonemes, but it’s returning the wrong phonemes in my language when the word contains double n, for example “madonna”.

I wanted it to treat such words as “madona”.

I tried to create an extra file dictionary, compile it and use but it doesn’t matter rule I try to create, it doesn’t work.

What I’m doing now is basically creating a list of all the possible words I can think of with double Ns and map them to a single N but it’s not the best solution… I just wanted to fix it in a single dictionary rule.

thanks!