speechio / chinese_text_normalization Goto Github PK

Hey, I really love your work here. It is super helpful for what I'm working on right now.

Are you guys planning to release a PyPI package of this in the near future?

Hi, thanks for great work. But when I test run_cn.sh, seems it can't process "万" and "点", for example,

Google use thrax and sparrowhawk to build TN system. Since thrax is already used in ITN, is it handy to add corresponding thrax rules in TN？

As I know, src/en/verbalizer is for TN. How to get English ITN.

Thank you.

First of all: Thanks for your work, this project is so awesome!

I just have one question:
What is the license of this project?

Thanks!

input date-time likes:

2022-03-03
2022/03/03

got output:

零三分之两千零二十二 零三
二零二二 零三 零三

“三分之二”转到“3/2”很容易做到，但是如何做到“三分之二”转到“2/3”？ @xingchensong @dophist

比如：3/5 表达5分之三，3米/秒表达3米每秒，4/2表达4除以2.类似这种怎么表达？

Hi
If I donot want to convert idiom such as “三七二十一”，what should I do to support this?

thanks

我看这个项目是进行文本的规范化处理，而不是，针对语音识别中的部分错误进行纠错的吗？