add new voices to freeTTS other than us voices

Question

I am working on a speech recognition project, therefore I need to have audio files for more than thousands of words. I have done everything correctly in order to use Mbrola voices in FreeTTS but I can only use us1,us2,us3. It doesn't recognize any other voices.

When I try to use mbrola_en1 I get this error:

Invalid Voice: Mbrola_en1
valid voices: alan kevin kevin16 mbrola_us1 mbrola_us2 mbrola_us3

I read all the questions related to this but unfortunately none of them are answered.

Does anyone know how to fix this issue?

Answer 1

The MbrolaVoiceDirectory class is hardcoded for us1, us2, us3. I copied the source code of the file to my own project, pointed the freetts.voices environment variable at it, and gently hacked it to look for the other filenames.