How shravasti221/birminghack works
Project type and purpose
• StoryPhonic is a full-stack web application (built with Python/Flask) that converts user-supplied text into a narrated audio experience using AI-driven text parsing, voice assignment, and speech synthesis. The README confirms features such as dynamic dialogue extraction, celebrity voice casting, and exportable audio.
File structure
Top-level components:
• app.py & routes.py – Flask application and HTTP route definitions
• config.py – configuration for API keys, voice model endpoints, etc.
• utils.py, load_text.py, voice_assigner.py, merge_audio_files.py, get_voices.py – core processing modules
• generator_async_SSE.py – Server-Sent Events stream for real-time audio generation
• audio_samples/ & voices.csv / chars_gender.txt – static data stores for sample audio, voice metadata, and parsing dictionaries
• templates/index.html + static/style.css – front-end presentation layer
• requirements.txt – Python dependencies
README for architectural hints
• The app ingests raw text (user upload or paste).
• It extracts characters and dialogues, then assigns user-chosen celebrity voices.
• It invokes an external AI speech synthesis API for each turn of dialogue or narration.
• It merges the resulting audio clips into a final output (a downloadable audio file).