Spaces:
Running
title: ConvoTrack
emoji: 📚
colorFrom: indigo
colorTo: yellow
sdk: docker
pinned: false
license: mit
short_description: 'A chatbot for Minutes of Meeting '
NOTE: Currently, the application supports only JSON files. If you want to modify it for other file types, update the following line in the code: uploaded_file = st.file_uploader("Upload a JSON file for processing", type=["json"])
For example, to support CSV files: uploaded_file = st.file_uploader("Upload a file for processing", type=["csv"])
To support multiple file types (e.g., JSON, CSV, TXT), you can modify the code as follows: uploaded_file = st.file_uploader("Upload a file for processing", type=["json", "csv", "txt"])
USER FLOW:
1]Upload JSON File: The app processes the uploaded file and extracts meeting passages.
2]Embedding Generation: The passages are converted into embeddings using a pre-trained transformer model.
3]FAISS Indexing: Embeddings are stored in a FAISS index for efficient retrieval.
4]User Query: Users input questions about the document.
5]Retrieval & Reranking: Relevant passages are retrieved and reranked for relevance.
6]Response Generation: Contextual answers are generated using Google Generative AI.