Back to datasets
Dataset assetOpen Source CommunityAnimeAI Training
ACCD
Aiming to promote Chinese AI character creation, this project continuously collects typical anime character dialogue data (ACCD) and stores it in a public repository, which can be used for AI character training or literary creation learning.
Source
github
Created
Jun 11, 2023
Updated
Jun 11, 2023
Signals
456 views
Availability
Linked source ready
Overview
Dataset description and usage context
Dataset Overview
Dataset Purpose
The project aims to foster Chinese AI character creation by collecting and providing typical anime character dialogue data (ACCD) for AI character training or literary creation learning.
Data Source and Usage Restrictions
- Data are sourced from publicly available anime series, intended solely for AI learning; commercial use is prohibited.
- The content does not represent the views of the project team, and the team does not guarantee the authenticity of the content.
Data Storage and Format
- Data are stored as JSON files, using the three keys instruction, input, output, and extending with MBTI personality parameters.
- Data are divided into three categories: World Knowledge, Self‑Recognition, and Dialogue, with the dialogue portion optionally merged as needed.
Data Application Tools
- excel2json tool: converts dialogues from Excel into the project's standard JSON format.
Need downstream help?
Pair the dataset with AI analysis and content workflows.
Once the source passes your review, move straight into summarization, transformation, report drafting, or presentation generation with the JuheAI toolchain.