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.
76 lines
1.7 KiB
Python
76 lines
1.7 KiB
Python
from __future__ import annotations
|
|
|
|
import tempfile
|
|
from pathlib import Path
|
|
|
|
import pytest
|
|
|
|
from umbraaudio.analysis import CharacterProfile
|
|
from umbraaudio.assembly import AudioStitcher
|
|
|
|
|
|
@pytest.fixture
|
|
def temp_dir():
|
|
with tempfile.TemporaryDirectory() as d:
|
|
yield Path(d)
|
|
|
|
|
|
@pytest.fixture
|
|
def sample_wavs(temp_dir: Path) -> list[Path]:
|
|
import numpy as np
|
|
import soundfile as sf
|
|
|
|
paths = []
|
|
sr = 24000
|
|
for i in range(3):
|
|
path = temp_dir / f"test_{i}.wav"
|
|
audio = np.random.randn(sr).astype(np.float32) * 0.1
|
|
sf.write(str(path), audio, sr)
|
|
paths.append(path)
|
|
return paths
|
|
|
|
|
|
@pytest.fixture
|
|
def stitcher() -> AudioStitcher:
|
|
return AudioStitcher(
|
|
silence_between_utterances_ms=100,
|
|
silence_between_paragraphs_ms=200,
|
|
silence_between_chapters_ms=300,
|
|
sample_rate=24000,
|
|
)
|
|
|
|
|
|
@pytest.fixture
|
|
def narrator_profile() -> CharacterProfile:
|
|
return CharacterProfile(
|
|
name="Narrator",
|
|
gender="male",
|
|
age_range="adult",
|
|
is_narrator=True,
|
|
)
|
|
|
|
|
|
@pytest.fixture
|
|
def character_profile() -> CharacterProfile:
|
|
return CharacterProfile(
|
|
name="John Smith",
|
|
aliases=["John", "Smithy"],
|
|
gender="male",
|
|
age_range="young_adult",
|
|
pitch_register="low",
|
|
energy="measured",
|
|
timbre="warm",
|
|
texture="smooth",
|
|
formality="formal",
|
|
emotional_baseline="neutral_calm",
|
|
pace_baseline="slow",
|
|
distinguishing_trait="deep resonant voice",
|
|
accent_marker="british_rp",
|
|
voice_description="A deep, warm voice with a British RP accent",
|
|
)
|
|
|
|
|
|
@pytest.fixture
|
|
def registry_db_path(temp_dir: Path) -> Path:
|
|
return temp_dir / "voice_registry.db"
|