c86e323d7a
Quality-first multi-pass audiobook generation from English manuscripts. Includes: - umbraaudio/ core package (pipeline, config, voices, web review UI) - scripts/ CLI utilities + model download script - tts_server/ remote TTS server - specs/ design docs, tests/, pyproject.toml + uv.lock - Qwen3-TTS git submodule (pinned 022e286) with local demo.py UI patch saved at scripts/qwen3_tts_demo_umbra_patch.patch - .env with runtime config/secrets Excluded (see .gitignore): models/ (downloaded via scripts/download_models.sh), output/, logs/, .venv, .umbra runtime data, and book projects (DealHunterShip_*), which are committed separately.
24 lines
366 B
Plaintext
24 lines
366 B
Plaintext
# Qwen3-TTS server runtime for the UmbraAudio pipeline
|
|
# Python 3.12, Linux/CUDA (GPU) target.
|
|
#
|
|
# On Linux, install a CUDA build of PyTorch FIRST (see install_server.sh),
|
|
# then this file.
|
|
|
|
qwen-tts==0.1.1
|
|
transformers==4.57.3
|
|
accelerate==1.12.0
|
|
gradio
|
|
librosa
|
|
torchaudio
|
|
soundfile
|
|
sox
|
|
onnxruntime
|
|
einops
|
|
# Server runtime
|
|
fastapi
|
|
uvicorn
|
|
pydantic
|
|
anyio
|
|
httpx
|
|
numpy
|