Spaces:
Running
Running
NaikPriyank
commited on
Update README.md
Browse filesFeatures
π JSON File Support: Upload a JSON file with meeting data for processing.
π Intelligent Search: Retrieve the most relevant passages from the document using FAISS.
π§ Contextual Reranking: Use a ranking model to rerank retrieved passages for improved accuracy.
π¬ Interactive Chat Interface: Ask questions about the meeting content and get answers generated by Generative AI.
π Memory Management: Keep track of chat history using a custom memory module.
README.md
CHANGED
@@ -18,3 +18,16 @@ uploaded_file = st.file_uploader("Upload a file for processing", type=["csv"])
|
|
18 |
|
19 |
To support multiple file types (e.g., JSON, CSV, TXT), you can modify the code as follows:
|
20 |
uploaded_file = st.file_uploader("Upload a file for processing", type=["json", "csv", "txt"])
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
18 |
|
19 |
To support multiple file types (e.g., JSON, CSV, TXT), you can modify the code as follows:
|
20 |
uploaded_file = st.file_uploader("Upload a file for processing", type=["json", "csv", "txt"])
|
21 |
+
|
22 |
+
USER FLOW:
|
23 |
+
1]Upload JSON File: The app processes the uploaded file and extracts meeting passages.
|
24 |
+
|
25 |
+
2]Embedding Generation: The passages are converted into embeddings using a pre-trained transformer model.
|
26 |
+
|
27 |
+
3]FAISS Indexing: Embeddings are stored in a FAISS index for efficient retrieval.
|
28 |
+
|
29 |
+
4]User Query: Users input questions about the document.
|
30 |
+
|
31 |
+
5]Retrieval & Reranking: Relevant passages are retrieved and reranked for relevance.
|
32 |
+
|
33 |
+
6]Response Generation: Contextual answers are generated using Google Generative AI.
|