Turning Viewer Voices into Actionable Insights
To automatically analyze YouTube video comments and generate structured summaries highlighting audience sentiment, key feedback themes, and actionable insights for content creators.
Content creators waste hours manually reading comments to understand:
☑️ What resonated with viewers
☑️ What criticisms matter most
☑️ What improvements audiences request
(92% of creators cite this as their top time drain - Adobe Creativity Study)
- Growing Creators (1K-100K subs) who lack teams to analyze comments
- Educational/Tutorial Channels needing precise feedback on content clarity
- Small Businesses using YouTube for customer engagement
- Podcasters repurposing video content
| Component | Function |
|---|---|
| Comment Miner | Fetches top 100 comments via YouTube API |
| Sentiment Radar | Classifies praise/criticism with VADER NLP |
| Topic Spotter | Groups feedback into themes (e.g. "Audio Issues", "Pacing") |
| Insight Generator | Creates executive summary using GPT-5 |
| Report Builder | Outputs 1-page PDF with key metrics & recommendations |
"Transform 100+ scattered comments into a 30-second readable report highlighting what truly matters to your audience."
✓ Cost-free for creators (unlike enterprise tools like Hootsuite)
✓ Sarcasm-aware analysis (unlike YouTube Studio's keyword filters)
✓ Action-oriented summaries (e.g. "3 viewers requested chapters at 5:32")
Why This Matters:
Creators using such tools see 40% faster content iteration (TubeBuddy case studies). This project democratizes audience analytics for those who can't afford $50+/mo SaaS tools.
/backend
│
├── app/
│ ├── main.py # FastAPI app entrypoint
│ ├── config.py # Configurations and API keys (YouTube, OpenAI)
│ ├── api/
│ │ ├── __init__.py
│ │ ├── comment_miner.py # Step 1: YouTube API Integration
│ │ ├── sentiment_radar.py # Step 2: Sentiment analysis & sarcasm heuristics
│ │ ├── topic_spotter.py # Step 3: Clustering & theme detection
│ │ ├── insight_generator.py # Step 4: GPT-5 prompt & summary logic
│ │ ├── report_builder.py # Step 5: PDF generation utilities
│ │ └── router.py # API router combining all routes
│ │
│ ├── models/ # Data models and schemas (Pydantic)
│ │ ├── comment.py # Comment data schema
│ │ ├── sentiment.py # Sentiment classification schema
│ │ └── report.py # Report data structure
│ │
│ ├── services/ # Business logic helpers
│ │ ├── youtube_service.py # YouTube API client and helpers
│ │ ├── nlp_service.py # NLP processing (VADER + clustering)
│ │ └── openai_service.py # OpenAI API wrapper and prompt handlers
│ │
│ └── utils/ # Utility functions (sanitization, logging, errors)
│ └── helpers.py
│
├── tests/ # Unit and integration tests per module
│
├── requirements.txt # Dependencies
└── Dockerfile # Container setup (optional for deployment)