Hi Jesin,
Yes, AuToBI has been used to train models in other languages. The biggest issue can be with how different ToBI standards define the inventory of tones. I've trained models for German, Italian, French and Mandarin. There was an effort by Helena Moniz and others to extend this to Portuguese. I think David Escudero Mancebo did some work on Spanish -- though I'm not sure if in the end he developed his own tool or modified AuToBI. While there are citations to results in these directions, i'm not sure that the papers describing results will be useful in giving guidance in how to modify the code to support different annotation standards and other potential pitfalls.
I'm happy to give you some insight about how to develop this. It might be more useful to take this offline and email directly.
-Andrew