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.
8 lines
376 B
Bash
Executable File
8 lines
376 B
Bash
Executable File
#!/bin/bash
|
|
# Qwen3-TTS 1.7B-Base Gradio server launcher (all paths on the 1TB drive)
|
|
export HF_HOME=/Volumes/1TB_APL_Storage/UmbraAudio/.hf_home
|
|
export PATH=/Volumes/1TB_APL_Storage/UmbraAudio/.venv/bin:$PATH
|
|
exec qwen-tts-demo /Volumes/1TB_APL_Storage/UmbraAudio/models/Qwen3-TTS-12Hz-1.7B-Base \
|
|
--device mps --no-flash-attn --dtype bfloat16 \
|
|
--ip 0.0.0.0 --port 8000
|