omnivoice/pyproject.toml

Rama main · réplica de lectura del repositorio original.

[build-system]
requires = ["hatchling"]
build-backend = "hatchling.build"

[project]
name = "geny-omnivoice"
version = "0.1.0"
description = "Geny's FastAPI wrapper around a vendored snapshot of k2-fsa/OmniVoice for in-cluster TTS."
readme = "README.md"
license = "Apache-2.0"
requires-python = ">=3.10"
authors = [{ name = "Geny" }]
keywords = ["tts", "omnivoice", "geny"]
classifiers = [
    "Programming Language :: Python :: 3",
    "Topic :: Multimedia :: Sound/Audio :: Speech",
    "Operating System :: OS Independent",
]

# NOTE: torch / torchaudio are installed separately in the Dockerfile with
# the matching CUDA wheel index. They are intentionally NOT pinned here so
# that local CPU/MPS development can pick whichever wheel is convenient.
dependencies = [
    "torch>=2.4",
    "torchaudio>=2.4",
    "transformers>=5.3.0",
    "accelerate",
    "huggingface_hub",
    "pydub",
    "tensorboardX",
    "numpy",
    "soundfile",
    "librosa",
    "fastapi>=0.115",
    "uvicorn[standard]>=0.32",
    "python-multipart",
    "httpx",
    "pydantic>=2.0",
    "pydantic-settings>=2.0",
]

[project.optional-dependencies]
dev = [
    "pytest>=8.0",
    "pytest-asyncio>=0.23",
    "ruff",
]

[project.scripts]
geny-omnivoice-server = "server.main:run"

[tool.hatch.build.targets.wheel]
packages = ["omnivoice_core", "server"]