Cyril Dupland
feat transcription: implement audio file chunking for transcription service to handle files exceeding 25 MB limit. Add pydub dependency for audio processing and enhance transcribe method to manage large audio files by splitting them into smaller segments. Update documentation for clarity on new functionality.
d7bed1d | # FastAPI et serveur | |
| fastapi>=0.115.6,<0.128.0 | |
| uvicorn[standard]>=0.32.0,<1 | |
| python-multipart>=0.0.9,<1 | |
| # Validation et configuration | |
| pydantic>=2.10.6,<3 | |
| pydantic-settings>=2.10.1,<3 | |
| python-dotenv>=1.0.1,<2 | |
| # Securite JWT | |
| python-jose[cryptography]>=3.3.0,<4 | |
| passlib[bcrypt]>=1.7.4,<2 | |
| # LangChain et IA | |
| langchain>=0.3.27,<0.4 | |
| langchain-openai>=0.3.9,<0.4 | |
| langchain-community>=0.3.20,<0.4 | |
| langchain-mistralai>=0.2.4,<0.3 | |
| mistralai>=1.0.0,<2 | |
| requests>=2.32.0,<3 | |
| httpx>=0.27.0,<1 | |
| aiohttp>=3.11.0,<4 | |
| langgraph>=0.2.35,<0.3 | |
| openai>=1.74.0,<2 | |
| # Supabase | |
| supabase>=2.7.4,<3 | |
| vecs>=0.4.0,<1 | |
| # Pipecat Voice (WebRTC + STT + TTS + VAD) | |
| # Pin >=0.0.90 for webrtc/silero extras. | |
| pipecat-ai[silero,openai,webrtc]>=0.0.90 | |
| # Daily transport: Linux-only (daily-python requires glibc); skipped on Windows. | |
| daily-python>=0.10.0 ; sys_platform == "linux" | |
| pydub>=0.25.1,<1 | |
| # Summarizer tooling | |
| reportlab>=4.4.4,<5 | |
| markdown>=3.7,<4 | |
| markdown-pdf>=1.13.0 | |
| # Ecologits Carbon Footprint | |
| ecologits>=0.8.2,<1 | |
| # Tests | |
| pytest>=8.0.0,<9 | |