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.
45 lines
1.2 KiB
Bash
45 lines
1.2 KiB
Bash
# UmbraAudio Environment Configuration
|
|
# Copy this to .env and modify as needed
|
|
|
|
# --- Project ---
|
|
UMBRA_PROJECT_DIR=.
|
|
UMBRA_DATA_DIR=.umbra
|
|
|
|
# --- TTS ---
|
|
UMBRA_TTS__ENGINE=qwen
|
|
UMBRA_TTS__DEVICE=mps
|
|
UMBRA_TTS__SAMPLE_RATE=24000
|
|
UMBRA_TTS__MAX_RETRIES=3
|
|
UMBRA_TTS__RETRY_DELAY=2.0
|
|
UMBRA_TTS__VOICEDESIGN_MODEL=/Volumes/1TB_APL_Storage/UmbraAudio/models/Qwen3-TTS-12Hz-1.7B-VoiceDesign
|
|
UMBRA_TTS__BASE_MODEL=/Volumes/1TB_APL_Storage/UmbraAudio/models/Qwen3-TTS-12Hz-1.7B-Base
|
|
|
|
# --- LLM ---
|
|
UMBRA_LLM__PROVIDER=auto
|
|
UMBRA_LLM__LMSTUDIO_BASE_URL=http://localhost:1234/v1
|
|
UMBRA_LLM__LMSTUDIO_MODEL=qwen/qwen3.5-9b
|
|
UMBRA_LLM__OPENAI_MODEL=gpt-4o
|
|
UMBRA_LLM__TEMPERATURE=0.3
|
|
UMBRA_LLM__MAX_TOKENS=4096
|
|
|
|
# --- Web ---
|
|
UMBRA_WEB__HOST=0.0.0.0
|
|
UMBRA_WEB__PORT=8420
|
|
UMBRA_WEB__AUTO_OPEN=false
|
|
|
|
# --- Quality ---
|
|
UMBRA_QUALITY__MIN_MOS=3.8
|
|
UMBRA_QUALITY__MIN_TEST_MOS=4.0
|
|
UMBRA_QUALITY__MAX_WER=0.03
|
|
UMBRA_QUALITY__MIN_SPEAKER_SIMILARITY=0.80
|
|
UMBRA_QUALITY__MIN_TEST_SPEAKER_SIMILARITY=0.85
|
|
UMBRA_QUALITY__MIN_ARTIFACT_SCORE=3.5
|
|
|
|
# --- Audio ---
|
|
UMBRA_AUDIO__TARGET_RMS_DB=-20.0
|
|
UMBRA_AUDIO__PEAK_LIMIT_DB=-3.0
|
|
UMBRA_AUDIO__NOISE_FLOOR_DB=-60.0
|
|
UMBRA_AUDIO__SILENCE_BETWEEN_UTTERANCES_MS=500
|
|
UMBRA_AUDIO__SILENCE_BETWEEN_PARAGRAPHS_MS=2000
|
|
UMBRA_AUDIO__SILENCE_BETWEEN_CHAPTERS_MS=4000
|