Tired of spending endless hours creating dull presentations? Wishing for a magical tool that can generate stunning PPTs for you in seconds? Well, fret no more! Introducing Auto_PPT!
🔥 中文指南
中文用户请移步 中文指南.
Utilizing gpt-3.5-turbo and pptx, Auto_PPT effortlessly generates PPTX files with specified themes.
Below is an unmodified example generated by the project:
By starring the project, you demonstrate your recognition and help us gain more attention in the community.
This motivates us to continuously improve and develop new features to enhance your experience with Auto_PPT.
"Special thanks to Miraitowa-wsy for their sponsorship."
🌟 No more hassle: Simply enter the title, and Auto_PPT will instantly create a brand new PPTX for you without any extra effort!
🎩 The magic behind: We leverage the powerful gpt-3.5-turbo-16k interface to ensure stable and impressive PPT outlines with every generation.
💡 Creative use of md format: We uniquely utilize the md format in a multi-step chain to generate PPT text, making PPTX creation easier and more stable. Say goodbye to formatting hassles and focus on content creation!
🔗 Optimized and refactored using langChain in v1.0: Thanks to langChain, the code becomes simple, easy, and aesthetically pleasing!
🖼️ Scenic illustrations: We collaborate with Unsplash to provide the most exquisite illustrations, instantly adding vitality and aesthetics to your PPT.
🔒 Secure local deployment: If you're concerned about data security, fret not! Auto_PPT supports local deployment; simply add your OpenAI API key and Unsplash API key information.
The project operation requires a Python environment, and it is recommended to use Python 3 or above. The author uses Python 3.9 1 Creating a virtual environment
python - m venv venv2 Activate virtual environment
. venv/bin/activate3 Install required Python components
pip install - r requirements. txt4 Add your API key in config.ini 5 Modify/ The base absolute path of readconfig/mycofig.py makes it the folder path of config.ini 6 Run Project Run
python application.pyAlternatively (in production mode), the following commands need to be run in a Linux like environment
gunicorn - b 0.0.0.0:5000-- log level=debug -- threads 4 app: application>gunicorn. log 2>&1&7 Access http://127.0.0.1:5000
2023/7/3 | v0.5.1 | Birth of an idea | Completed ✔️
| Blueprint | Existing Issues | Completion |
|---|---|---|
| Deploy Online Service | UI is too rudimentary | Completed |
| Optimize Generation Format | Format is too monotonous | Adjusted paragraph spacing |
| Optimize Generation Speed | OpenAI API response is slow | Optimized service startup |
2023/7/6 | v1.0 | Refactor with langChain | Completed ✔
| Blueprint | Existing Issues | Completion |
|---|---|---|
| Optimize Generation Content | Generated content not detailed and accurate enough | Deferred to the next version |
| Optimize Generation Steps | Single step is not enough for a high-quality PPT | Completed on 7/14 |
| Use langChain to optimize the project | Refactor into a chain call | Completed on 7/14 |
2023/7/15 | v1.5 | Next version tasks | In progress 🧭
| Blueprint | Existing Issues | Completion |
|---|---|---|
| Support more md formats | Large amount of work for md formats | Just started |
| Refactor front-end code using a specific language | Lack of familiarity with front-end for backend engineers | Completed ✔ |
| Optimize theme effects | Lack of aesthetic ppt template reference | Just started |