tugstugi / mongolian-nlp Goto Github PK

ta sainu?

Thank you for providing this amazing GitHub repository for Mongolian NLP.

I am wondering how can i use other code to make an image to bicheg?
specifically, I am wondering about how can i encode the Mongolian code and train it using deep learning.
can you share your experience of how you build your dataset and training processes? thank you.

Dear @tugstugi,

Recently, I have released multiple morphological resources. It would be so nice if you link them from here. :)
I believe they are useful for the community. Here are the links.

• Inflectional morphology - https://github.com/unimorph/khk
• Derivational morphology - https://github.com/kbatsuren/MorphyNet

Best regards,
Huygaa

Hi thereh,

The dataset (Mongolian POS dataset of the Mongolian National University) cannot be downloaded from the following link.
http://www.panl10n.net/center-for-research-on-language-processing-crlp-national-university-of-mongolia-mongolia/

Can you give me a link where I can download the dataset?

Thank you.

Hi @tugstugi 👋

Firstly, thanks for contributing so much to open speech tech for Mongolian!

You might add a link to the Mongolian model in the Coqui Model Zoo https://coqui.ai/mongolian/itml/v0.1.1

The model zoo is constantly growing, and if you'd like to train any Mongolian STT or TTS models, we'd love to put them in the zoo!

Feel free to chat with us in our chat room :) https://gitter.im/coqui-ai/STT

Thanks for posting these amazing resources! These are a really great place to get started. In the next year I am really interested in getting sentiment analysis and NER working in Mongolian. Is the biggest problem labeled datasets currently?