Dose it support Chinese song?

[yangkang2318]

Hello,this project is amazing!But dose it support Chinese song?

[mathigatti]

Hey! No, but it's a good idea, I just added that feature :) Please try it and let me know how that goes

[yangkang2318]

There is a bug in line 33,and occurs TypeError: 'int' object is not callable.Because when run in python,the "voice_index" is identified as "int" object instead of "voice_index" method.

[yangkang2318]

And you should modify line 26 in "__main__.py"，the last args should be "args["language"]"

[yangkang2318]

It should be chinese instead of mandarin

Beacuse in your code is "chinese"

so there should be "chinese".

[mathigatti]

Oh sorry and thanks for reviewing it, I just pushed some possible fixes I will test it thoroughly tomorrow

[yangkang2318]

When using the Chinese song ，the performance is worse.Maybe the file"lyrics_tokenizer.py" caused the results.
Because chinese words doesn't need to tokenize into subwords

[mathigatti]

ah right, I don't know much about japanese and mandarin. If you can send me a text example and how it should be splitted into notes that would be great.

For example in english, if I have this sentence "hello this is my song" and I have 6 notes it's splitted into something like this: ["he", "llo", "this", "is", "my", "song"], then I assign each text piece to each note.

[mathigatti]

Do you have some text to try? I added a basic tokenization, given some text like this: 曲项向天歌 it assigns each character to a different note

"曲项向天歌" is converted into this [u'\u66f2', u'\u9879', u'\u5411', u'\u5929', u'\u6b4c']

[yangkang2318]

I fix this problem，because the musescore decode the midi into .xml is wrong.
Thanks for your help and project!

[mathigatti]

Awesome! Do you want to share how you fixed it?