# chatppt **Repository Path**: wshmaple/chatppt ## Basic Information - **Project Name**: chatppt - **Description**: No description available - **Primary Language**: Unknown - **License**: Not specified - **Default Branch**: main - **Homepage**: None - **GVP Project**: No ## Statistics - **Stars**: 0 - **Forks**: 0 - **Created**: 2025-04-24 - **Last Updated**: 2025-04-24 ## Categories & Tags **Categories**: Uncategorized **Tags**: None ## README # ChatPPT ChatPPT is a tool powered by chatgpt/ollama that helps you generate PPT/slide. It supports output in English and Chinese. ## Table of Contents - [What's New](#whats-new) - [What is ChatPPT](#what-is-chatppt) - [Requirements](#requirements) - [Installation](#installation) - [Usage](#usage) - [Contributing](#contributing) - [License](#license) ## What's New ChatPPT now supports Ollama and includes a sample UI. ![UI demo 1](ui_demo_1.png) ![UI demo 2](ui_demo_2.png) ## What is ChatPPT ChatPPT is powered by chatgpt/ollama. It can help you generate PPT/slide in English and Chinese. ![What is GPT | 600](demo1.png) ![什么是AWS | 400](demo2.png) ## Requirements Python 3.8.10 or higher ## Installation ### Ollama Follow the [guide](https://ollama.com/) to install ollama ### OpenAI Generate your OpenAI API key at ## Usage 1. Install requirements ``` pip install -r requirements.txt ``` 2. Start Streamlit ``` streamlit run chatppt_ui.py ``` 3. Open the Streamlit URL in your browser () ![UI](ui.png) > You can also use ChatPPT in the command line: ```bash > python chatppt.py -h usage: chatppt.py [-h] [-m {openai,ollama}] -t TOPIC [-k API_KEY] [-u OLLAMA_URL] [-o OLLAMA_MODEL] [-p PAGES] [-l {cn,en}] I am your PPT assistant, I can help to you generate PPT. options: -h, --help show this help message and exit -m {openai,ollama}, --ai_model {openai,ollama} Select the AI model -t TOPIC, --topic TOPIC Your topic name -k API_KEY, --api_key API_KEY Your api key file path -u OLLAMA_URL, --ollama_url OLLAMA_URL Your ollama url -o OLLAMA_MODEL, --ollama_model OLLAMA_MODEL Specify the Ollama model to use -p PAGES, --pages PAGES How many slides to generate -l {cn,en}, --language {cn,en} Output language