requirements.txt

# -----------------------------------------------------------------------------
# Sahel-Agri Voice AI — HuggingFace Spaces (cpu-basic) runtime dependencies
#
# RULES:
#   - Do NOT pin torch/torchaudio — Spaces pre-installs the correct CUDA build.
#     Re-pinning them forces a 1 GB re-download on every build.
#   - Do NOT add training-only packages (peft, evaluate, bitsandbytes,
#     audiomentations) — they are only needed in the Kaggle notebook.
#   - Do NOT add dev tools (pytest, pytest-asyncio) — not run in production.
#   - f5-tts is intentionally omitted — it is a ~2 GB install and is only
#     needed for GPU voice cloning.  src/tts/f5_tts.py falls back to MMS-TTS
#     gracefully when the package is absent.
#   - speechbrain is intentionally omitted — it is heavy and speaker-id is
#     not on the critical path at runtime.
# -----------------------------------------------------------------------------

# HuggingFace core
transformers==5.5.0
datasets==4.8.4
accelerate==1.13.0
huggingface-hub==1.9.0

# Audio processing
librosa==0.10.2
soundfile==0.12.1

# Metrics (used in app.py CER display)
jiwer==3.0.4

# Config & environment
pyyaml==6.0.2
python-dotenv==1.1.0

# Gradio version is controlled by sdk_version in README.md — do not pin here
# to avoid conflict with the version HF Spaces pre-installs.

# Pydantic v2
pydantic==2.11.3

# Utilities
numpy==2.2.4
scipy==1.15.2
rapidfuzz==3.13.0

# Kaggle API (used by Self-Teaching tab to trigger training runs)
kaggle>=1.6.0

# Document parsing for Knowledge Base tab (PDF/Word/TXT upload → vocabulary.jsonl)
pypdf>=4.0.0
python-docx>=1.1.0