InsightEngine is a versatile web application built with Streamlit, providing various functionalities such as chatbot, image captioning, sentiment analysis, language translation, topic modeling, YouTube transcriber, and document summarizer.
Check out the live demo here.
- ChatBot: Interactive chat interface powered by Gemini Pro models for natural language understanding and generation.
- Image Captioning: Generate captions for uploaded images using Gemini Pro Vision models.
- Sentiment Analysis: Analyze sentiment of provided text with Vader Sentiment Analysis from NLTK.
- Language Translation: Translate text between various languages using Google Translate API.
- Topic Modeling: Discover key themes within text data using Latent Dirichlet Allocation (LDA).
- YouTube Transcriber: Transcribe YouTube videos into text with language selection and paragraph segmentation.
- Document Summarizer: Summarize text documents to extract key information efficiently.
To run this project locally, follow these steps:
Clone the repository: git clone https://github.com/varnitsaxena7/InsightEngine.git
Install dependencies: pip install -r requirements.txt
Configure API Keys: Obtain a Google API key and add it to config.json.
Run the Streamlit app: streamlit run app.py
Open your browser and navigate to http://localhost:8501 to view the application.
Streamlit: Front-end framework for building interactive web applications.
NLTK: Natural Language Toolkit for sentiment analysis.
Googletrans: Google Translate API for language translation.
YouTube Transcript API: For fetching transcripts of YouTube videos.
PIL: Python Imaging Library for image processing.
Scikit-learn: For topic modeling using Latent Dirichlet Allocation (LDA).
Chardet: For automatic detection of text file encoding.