How insolve-wojciech-bednorz/artificial_voiceline_playback works
Project type and purpose
• This is a full-stack “voice-line management and radio playback” system.
• Frontend: a Next.js (React) SPA under nextjs-voiceline-app/ for adding/editing/searching/scheduling voice lines.
• Backend: Python scripts (main.py, voice_system.py, models.py, etc.) that handle data storage, TTS generation (ElevenLabs), scheduling, and radio streaming.
File structure
• Top level:
• audio_files/ – stored MP3 voice lines
• voice_lines.json – metadata DB for lines and schedules
• config.yaml – global settings (station parameters, scheduler rules, TTS voice settings)
• main.py & voice_system.py – core orchestrators (API endpoints? CLI? scheduler loop)
• models.py – data definitions (VoiceLine, Schedule, StationConfig)
• requirements.txt – Python dependencies (e.g., Flask/FastAPI? APScheduler?)
• Frontend (nextjs-voiceline-app/):
• src/app/ – pages (dashboard, settings)
• src/components/ – UI building blocks (modals, tables, settings panels)
• src/utils/api.ts – HTTP calls to backend
• src/utils/elevenlabs.ts – client wrapper (optional)
• Deployment/build files: package.json, next.config.ts, tailwind.config.js, build scripts (setup.sh/.bat).
README
• Confirms: “Program tworzący, zarządzający i odtwarzający linie głosowe wraz z radiem.”
• Key functions: create TTS lines, manage metadata & schedule, play back on a radio stream.