Dataset for Chatbot training.

1. Language requirements: Chinese, English, Spanish 2. Data magnitude: 100,000 groups for each language 3. Field requirements: retail, medical, finance. 4. Data type: dialogue and email, which accounts for 50% respectively. 5. Content requirements: text dialogue of customer service scenario and email data are required. The number of conversations is at least 3 Qs and 3 As, and email exchanges are at least 1 Q and 1 A. 6. Personal or corporate information that is mentioned in the dialogue or email can be anonymized.