baidu / Unit Dmkit
Licence: apache-2.0
Stars: ✭ 279
Projects that are alternatives of or similar to Unit Dmkit
Dstc8 Schema Guided Dialogue
The Schema-Guided Dialogue Dataset
Stars: ✭ 277 (-0.72%)
Mutual labels: dialogue, dialogue-systems
Whiskers
A Visual Dialogue Editor made using the Godot Engine
Stars: ✭ 229 (-17.92%)
Mutual labels: dialogue, dialogue-systems
Anyq
FAQ-based Question Answering System
Stars: ✭ 2,336 (+737.28%)
Mutual labels: dialogue, dialogue-systems
Dialog Generation Paper
A list of recent papers regarding dialogue generation
Stars: ✭ 265 (-5.02%)
Mutual labels: dialogue, dialogue-systems
ADEM
TOWARDS AN AUTOMATIC TURING TEST: LEARNING TO EVALUATE DIALOGUE RESPONSES
Stars: ✭ 25 (-91.04%)
Mutual labels: dialogue, dialogue-systems
Dstc7 End To End Conversation Modeling
Grounded conversational dataset for end-to-end conversational AI (official DSTC7 data)
Stars: ✭ 141 (-49.46%)
Mutual labels: dialogue, dialogue-systems
Meld
MELD: A Multimodal Multi-Party Dataset for Emotion Recognition in Conversation
Stars: ✭ 373 (+33.69%)
Mutual labels: dialogue, dialogue-systems
Botml
Powerful markup language for modern chatbots.
Stars: ✭ 98 (-64.87%)
Mutual labels: bot, dialogue
Dialogue Understanding
This repository contains PyTorch implementation for the baseline models from the paper Utterance-level Dialogue Understanding: An Empirical Study
Stars: ✭ 77 (-72.4%)
Mutual labels: dialogue, dialogue-systems
TalkerMakerDeluxe
A FOSS Branching Game Dialogue Editor
Stars: ✭ 90 (-67.74%)
Mutual labels: dialogue, dialogue-systems
Rnnlg
RNNLG is an open source benchmark toolkit for Natural Language Generation (NLG) in spoken dialogue system application domains. It is released by Tsung-Hsien (Shawn) Wen from Cambridge Dialogue Systems Group under Apache License 2.0.
Stars: ✭ 487 (+74.55%)
Mutual labels: dialogue, dialogue-systems
Tgen
Statistical NLG for spoken dialogue systems
Stars: ✭ 179 (-35.84%)
Mutual labels: dialogue, dialogue-systems
Multiwoz
Source code for end-to-end dialogue model from the MultiWOZ paper (Budzianowski et al. 2018, EMNLP)
Stars: ✭ 384 (+37.63%)
Mutual labels: dialogue, dialogue-systems
Convlab 2
ConvLab-2: An Open-Source Toolkit for Building, Evaluating, and Diagnosing Dialogue Systems
Stars: ✭ 196 (-29.75%)
Mutual labels: dialogue, dialogue-systems
Nndial
NNDial is an open source toolkit for building end-to-end trainable task-oriented dialogue models. It is released by Tsung-Hsien (Shawn) Wen from Cambridge Dialogue Systems Group under Apache License 2.0.
Stars: ✭ 332 (+19%)
Mutual labels: dialogue, dialogue-systems
Deeppavlov
An open source library for deep learning end-to-end dialog systems and chatbots.
Stars: ✭ 5,525 (+1880.29%)
Mutual labels: bot, dialogue-systems
DlgSystem
Dialogue Plugin System for Unreal Engine | 🪞 Mirror of https://bit.ly/DlgSource
Stars: ✭ 136 (-51.25%)
Mutual labels: dialogue, dialogue-systems
dialogue-datasets
collect the open dialog corpus and some useful data processing utils.
Stars: ✭ 24 (-91.4%)
Mutual labels: dialogue, dialogue-systems
DMKit
DMKit作为UNIT的开源对话管理模块，可以无缝对接UNIT的理解能力，并赋予开发者多状态的复杂对话流程管理能力，还可以低成本对接外部知识库，迅速丰富话术信息量。
快速开始
编译DMKit
DMKit基于brpc开发并提供HTTP服务，支持MacOS，Ubuntu，Centos等系统环境，推荐使用Ubuntu 16.04或CentOS 7。在编译DMKit之前，需要先安装依赖并下载编译brpc：
sh deps.sh [OS]
其中[OS]参数指定系统类型用于安装对应系统依赖，支持取值包括ubuntu、mac、centos。如果已手动安装依赖，则传入none。
使用cmake编译DMKit：
mkdir _build && cd _build && cmake .. && make
运行示例技能
DMKit提供了示例场景技能，在运行示例技能之前，需要在UNIT平台配置实现技能的理解能力：示例场景
根据UNIT平台创建的skill id修改编译产出_build目录下的conf/app/products.json文件，在其中配置所创建skill id与对应场景DMKit配置文件。例如，查询流量及续订场景，在UNIT平台创建skill id为12345，则对应的配置文件内容应为：
{
"default": {
"12345": {
"score": 1,
"conf_path": "conf/app/demo/cellular_data.json"
}
}
}
在_build目录下运行DMKit：
./dmkit
可以通过tools目录下的bot_emulator.py程序模拟与技能进行交互，使用方法为：
python bot_emulator.py [skill id] [access token]
更多文档
多语言支持
- PHP：PHP版本官方代码库
如何贡献
- 提交issue可以是新需求也可以是bug，也可以是对某一个问题的讨论。
- 对于issues中的问题欢迎贡献并发起pull request。
讨论
- 提issue发起问题讨论，如果是问题选择类型为问题即可。
- 欢迎加入UNIT QQ群（584835350）交流讨论。
Note that the project description data, including the texts, logos, images, and/or trademarks,
for each open source project belongs to its rightful owner.
If you wish to add or remove any projects, please contact us at [email protected].