dialogue-corpus
honkai_impact_3rd_chinese_dialogue_corpus
崩坏三游戏剧情语料
总计 92,421 句剧情对白（带有角色标签）+旁白，从崩坏3的“主线1黄昏、少女、战舰”到“主线第二部03间章：一个梦游者的苦痛”
本数据集从 honkai_impact_3rd_game_playthrough 视频数据集出发，经过 AI pipeline 最终获取结构化的文本剧情语料。
AI pipeline 概述如下：
分P下载视频（使用 BBDown 下载 BiliBili崩三剧情视频）
视频帧分割（每1秒取一帧画面）
逐帧 OCR 检测文本（使用 Paddle-OCR）
逐帧 VLM 结构化解析（使用 MiniCPM-V-2_6，输入为帧图像 + OCR结果，输出为结构化 JSON）
基于规则的后处理
规范化 VLM 输出（e.g., 去噪、排除格式有问题的输出）
中间帧的信息去重与归并（e.g.… See the full description on the dataset page: https://huggingface.co/datasets/mrzjy/honkai_impact_3rd_chinese_dialogue_corpus.ubuntu_dialogue_corpus_trainubuntu_dialogue_corpus_devtestmulti-relational-multi-party-chat-corpusMulti-Relational Multi-Party Chat Corpus (MRMP): Japanese text-based chats comprising first-time-meeting dialogues and family-included dialoguesAMRIT-Punjabi-Clinical-Dialogue-Corpus
ੴ AMRIT Punjabi Clinical Dialogue & Medical Diagnosis Corpus
☬ ਅੰਮ੍ਰਿਤ ਪੰਜਾਬੀ ਕਲੀਨਿਕਲ ਸੰਵਾਦ ਅਤੇ ਡਾਕਟਰੀ ਨਿਦਾਨ ਡਾਟਾਸੈੱਟ (v1.0)
👨💻 Research & Medical AI Architecture
Lead Developer: Gurpreet Singh Dhillon (Nam-toon Studio)
Mission: Free Autonomous AI Doctor for Humanity (ਦੁਨੀਆਂ ਦੇ ਲੋੜਵੰਦ ਲੋਕਾਂ ਲਈ ਮੁਫ਼ਤ AI ਡਾਕਟਰ)
Flagship Platform: AMRIT Research OS (100% Local Medical Intelligence)
📖 Dataset Overview / ਸੰਖੇਪ
The… See the full description on the dataset page: https://huggingface.co/datasets/Nam-toon-studio/AMRIT-Punjabi-Clinical-Dialogue-Corpus.Free_Dialogue_in_Saudi_Arabia_Corpus
Description
This dataset covers multiple scenarios such as banking, healthcare, insurance, sales, telecom, travel. The speakers are gender evenly, and each set of the audio is approximately 0.5 hour.
For more details, please refer to the link: https://dataoceanai.com/datasets/asr/free-dialogue-in-saudi-arabia-corpus/
Specification
ID:
King-ASR-919
SIZE:
113 hours
LANGUAGE:
Saudi Arabia
SPEAKERS:
100
AGES:
18-45 years old
DEVICES:
Mobile