How ojas-k070/poliemotifusion works
PoliEmotiFusion is a multimodal political-content emotion analysis platform. Its most complete user-to-result path is text submission, political-content screening, and seven-class emotion inference; image analysis has a political-content gate and experimental scene-emotion scoring, while audio supports multilingual transcription and translation. The UI also exposes video, history, insights, settings, and architecture views, though the README describes several as planned or design-stage capabilities. Some frontend/backend links and unsampled route internals are documented only at a high level, so those relationships are not asserted as verified calls.
Main entry points: main.py FastAPI Endpoints · pipeline.py Speech Pipeline · index.tsx Dashboard · translation_service.py Speech Translation
Components 22
User Interface
- Dashboard
frontend/src/routes/index.tsx - Text Analysis
frontend/src/routes/text.tsx - Image Analysis
frontend/src/routes/image.tsx - Audio Analysis
frontend/src/routes/audio.tsx - Video Analysis
frontend/src/routes/video.tsx - Analysis History
frontend/src/routes/history.tsx - Insights
frontend/src/routes/insights.tsx - Settings
frontend/src/routes/settings.tsx - Analysis Client
frontend/src/services/analysisService.ts
API Orchestration
- FastAPI Endpoints
backend/main.py - Request and Result Schemas
backend/schemas.py
Text and Image Analysis
- Text Political Gate
backend/political_service.py - Text Emotion Model
backend/model_service.py - Image Gate and Scoring
backend/image_model_service.py - Image Model Configuration
backend/image_config.py
Speech Intelligence
- Speech Pipeline
backend/speech_intelligence/pipeline.py - Audio Preprocessing
backend/speech_intelligence/preprocessing.py - Language Detection
backend/speech_intelligence/language_detection.py - Transcript Formatting
backend/speech_intelligence/transcript_formatter.py - Whisper Transcription
backend/speech_intelligence/whisper_service.py - Speech Translation
backend/speech_intelligence/translation_service.py
Other components
- Analyst
Connections 26
- Analyst → Dashboard: opens
README.md - Analyst → Text Analysis: submits text
README.md - Analyst → Image Analysis: uploads image
README.md - Analyst → Audio Analysis: uploads audio
README.md - Analyst → Video Analysis: opens module
README.md - Analyst → Analysis History: reviews analyses
README.md - Analyst → Insights: views summaries
README.md - Dashboard → Text Analysis: offers access
README.md - Dashboard → Image Analysis: offers access
README.md - Dashboard → Audio Analysis: offers access
README.md - Dashboard → Video Analysis: offers access
README.md - Audio Analysis → Analysis Client: calls
frontend/src/routes/audio.tsx - Analysis Client → FastAPI Endpoints: requests analysis
- FastAPI Endpoints → Request and Result Schemas: validates payloads
backend/main.py - FastAPI Endpoints → Text Political Gate: screens text
backend/main.py - FastAPI Endpoints → Text Emotion Model: classifies text
backend/main.py - FastAPI Endpoints → Image Gate and Scoring: scores image
backend/main.py - Image Gate and Scoring → Image Model Configuration: uses settings
backend/image_model_service.py - FastAPI Endpoints → Speech Pipeline: processes speech
backend/main.py - Speech Pipeline → Audio Preprocessing: preprocesses audio
backend/speech_intelligence/pipeline.py - Speech Pipeline → Language Detection: detects language
backend/speech_intelligence/pipeline.py - Speech Pipeline → Transcript Formatting: merges transcripts
backend/speech_intelligence/pipeline.py - Speech Pipeline → Whisper Transcription: transcribes chunks
backend/speech_intelligence/pipeline.py - Speech Pipeline → Speech Translation: translates speech
backend/speech_intelligence/pipeline.py - Speech Translation → Whisper Transcription: translates segments
backend/speech_intelligence/translation_service.py - Speech Translation → Transcript Formatting: merges translations
backend/speech_intelligence/translation_service.py