mio2333 / amadeus Goto Github PK

I downloaded the model from huggingface, and loaded it by calling espnet2.bin.tts_inference.
It can synthesize fluent Japanese, but it seem to be able to read only English letters, not words or sentences eg. "hello world"
I know this is possibly a problem with the tokenizer or encoders, but I don't know how to adjust or replace them.

I stumble across the Amadeus model on hugginface and I'm impressed by it. So I'd like to see the effect of doing further training, or training it on a different model. Furthermore, I'd like to expand the dataset with files from the other three STEINS;GATE games, beside just the first one. And though I'm able to extract the audio files, I can't seem to find a way to get the corresponding paired Japanese transcripts that you put in this repo (i.e. those in filelists/cris_train.txt). May I ask how to extract these files?