title | emoji | colorFrom | colorTo | sdk | python_version | sdk_version | app_file | pinned |
---|---|---|---|---|---|---|---|---|
Audio and Video Content Analyzer |
๐ฅ |
blue |
green |
streamlit |
3.8 |
1.27.2 |
app.py |
false |
This project is an all-in-one solution for audio and video content analysis:
- Summarization: Generates concise summaries using advanced Natural Language Processing, powered by HuggingFace Transformers.
- Semantic Retrieval: Enables you to find specific words, phrases, or segments, thanks to Whisper Timestamped by Linto.
- Chatbot Interface: Features a chatbot that can answer queries about the audio or video content, leveraging Language Models for Machines (LLM).
The project is built using Python and integrates various libraries including Whisper Timestamped, HuggingFace Transformers, and Streamlit for a seamless user experience.
.
โโโ Dockerfile # Dockerfile for setting up the environment
โโโ LICENSE # License file
โโโ README.md # This README file
โโโ YoutubeAudios # Directory containing YouTube audio files
โโโ _requirements.txt # Requirements file
โโโ app.py # Streamlit application file
โโโ config.py # Configuration file
โโโ keyword_retriever # Module for keyword retrieval
โ โโโ keyword_retreiver.py # Keyword retriever script
โโโ logger.py # Logging utility
โโโ notebooks # Jupyter notebooks for development and testing
โโโ pdf_test.py # PDF testing script
โโโ query_service # Query service module
โ โโโ query_engine.py # Query engine script
โโโ requirements.txt # Requirements file
โโโ resource_loader # Resource loader module
โ โโโ json_loader.py # JSON loader script
โ โโโ linkedin_loader.py # LinkedIn loader script
โ โโโ uploaded_video_loader.py # Uploaded video loader script
โ โโโ video_loader_interface.py# Video loader interface script
โ โโโ youtube_loader.py # YouTube loader script
โโโ summarization_service # Summarization service module
โ โโโ summarizer.py # Summarizer script
โโโ transcription_service # Transcription service module
โ โโโ transcriber.py # Transcriber script
โโโ utils.py # Utility functions
- Python 3.8+
- Docker (optional)
- Llama_index Framework for LLM application
- Whisper Timestamped - For semantic retrieval and timestamping
- HuggingFace Transformers - For summarization and NLP
- Streamlit - For the web interface
There are two methods to get the project up and running:
- Clone the repository.
- Navigate to the project directory.
- Build the Docker image:
docker build -t summarizer .
- Run the Docker container:
docker run -p 8501:8501 summarizer
- Open your web browser and go to
http://localhost:8501
.
- Clone the repository.
- Navigate to the project directory.
- Install the requirements:
pip install -r requirements.txt
- Run the Streamlit app:
streamlit run app.py
- Open your web browser and go to
http://localhost:8501
.
- Clone the repository.
- Navigate to the project directory.
- Install the requirements:
pip install -r requirements.txt
- Run the Streamlit app:
streamlit run app.py
- Open the Streamlit app in your web browser.
- Follow the instructions on the screen to upload or specify your audio/video content.
- Click "Submit" to generate a summary, perform semantic retrieval, or interact with the chatbot.
Please read CONTRIBUTING.md for details on our code of conduct, and the process for submitting pull requests.
This project is licensed under the MIT License - see the LICENSE.md file for details.