I am Joseph and am just starting out using Rasa to build a Chinese customer service chatbot for my company. I am stuck because I am not sure if I need to download any pretrained Chinese word vectors or can I just use my own conversation data, and how can I do it?
You can use your own data, however you will still need a different tokenizer for Chinese. You would also need a lot of data in order to make this better. I would recommend the Jieba-MITIE language model with the mitie pipeline.