How priyankshu-07/vision works
VISION takes YouTube URLs, PDFs, or typed text through a Streamlit interface, sends them to FastAPI for extraction and optional transcription, summarization, and quiz generation, then saves outputs for later history viewing. The graph follows the sampled call sites and README-described MongoDB/Groq capabilities; dashboard history reads MongoDB directly in the sampled UI code. Unsampled details are not inferred.
Main entry points: backend/main.py FastAPI routes · app.py Page router · frontend/main.py Input and results · dashboard.py History dashboard
Components 15
Learning interface
- Page router
frontend/app.py - Welcome page
frontend/home.py - Input and results
frontend/main.py - History dashboard
frontend/dashboard.py
Backend orchestration
- FastAPI routes
backend/main.py
Content processing
- PDF extraction
backend/pdf_reader.py - Audio transcription
backend/youtube_transcriber.py - Whisper model
models/whisper_model.py - T5 summarization
models/T5.py - Quiz generation
backend/question_generator.py - LLM prompt chain
backend/langchain_helper.py
History storage
- Persistence helper
backend/db.py - MongoDB results
Other components
- Learner
- Groq LLaMA API
Connections 16
- Learner → Page router: uses
README.md - Page router → Welcome page: routes to
frontend/app.py - Page router → Input and results: routes to
frontend/app.py - Page router → History dashboard: routes to
frontend/app.py - Input and results → FastAPI routes: submits inputs
frontend/main.py - FastAPI routes → PDF extraction: extracts text
backend/main.py - FastAPI routes → Audio transcription: transcribes URL
backend/main.py - Audio transcription → Whisper model: transcribes audio
backend/youtube_transcriber.py - FastAPI routes → T5 summarization: summarizes text
backend/main.py - FastAPI routes → Quiz generation: generates questions
backend/main.py - Quiz generation → LLM prompt chain: runs chain
backend/question_generator.py - LLM prompt chain → Groq LLaMA API: calls model
backend/langchain_helper.py - FastAPI routes → Persistence helper: saves results
backend/main.py - Persistence helper → MongoDB results: inserts records
backend/db.py - History dashboard → MongoDB results: reads history
frontend/dashboard.py - FastAPI routes → Input and results: returns outputs
frontend/main.py