LuxTTS Voice Cloning Backend (backends/luxtts_backend.py)

Async text-to-speech backend wrapping the LuxTTS (ZipVoice) model for zero-shot voice cloning with 48kHz output. Reuse it as a pluggable TTS engine with device selection, cached voice prompts, and model load progress.

# Source: https://github.com/mperkins0155/voicebox | License: MIT | Copyright 2026 Voicebox Contributors
# Provided via justclaw.it.com; keep this notice with the code.
"""
LuxTTS backend implementation.

Wraps the LuxTTS (ZipVoice) model for zero-shot voice cloning.
~1GB VRAM, 48kHz output, 150x realtime on CPU.
"""

import asyncio
import logging
from typing import Optional, Tuple

import numpy as np

from . import TTSBackend
from .base import (
    is_model_cached,
    get_torch_device,
    empty_device_cache,
    manual_seed,
    combine_voice_prompts as _combine_voice_prompts,
    model_load_progress,
)
from ..utils.cache import get_cache_key, get_cached_voice_prompt, cache_voice_prompt

logger = logging.getLogger(__name__)

# HuggingFace repo for model weight detection
LUXTTS_HF_REPO 

... (truncated -- full source via MCP)