LuxTTS

.env-example

@@ -0,0 +1,6 @@
+HF_CACHE_HOST_PATH=C:\users\joshua.bailey\.cache\huggingface
+HF_CACHE_CONTAINER_PATH=/home/user/app/.cache/huggingface
+HF_TOKEN=hf_...
+
+TORCH_CACHE_HOST_PATH=C:\users\joshua.bailey\.cache\torch
+TORCH_CACHE_CONTAINER_PATH=/home/user/app/.cache/torch

.gitignore

@@ -50,6 +50,8 @@ models/*
 config.json
 Qwen/
 .hf_cache/
+.cache/
+.env
 
 # Jupyter
 .ipynb_checkpoints/

Dockerfile

@@ -48,6 +48,8 @@ RUN pip install --no-cache-dir \
     torchvision==0.24.1+cu128
 RUN curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y
 ENV PATH="/home/user/.cargo/bin:${PATH}"
+WORKDIR /home/user/app
+COPY ./wheel /home/user/app/wheel
 RUN pip install --no-cache-dir \
     gradio==6.5.0 \
     soundfile==0.13.1 \
@@ -71,11 +73,24 @@ RUN pip install --no-cache-dir \
     onnxruntime \
     onnxruntime-gpu==1.23.2 \
     markdown==3.10.1 \
-    einops
+    einops \
+    "gradio_filelister @ file:wheel/gradio_filelister-0.4.0-py3-none-any.whl" \
+    lhotse \
+    safetensors \
+    tensorboard \
+    vocos \
+    pydub \
+    "transformers>=4.57.3,<5" \
+    cn2an \
+    inflect \
+    jieba \
+    pypinyin \
+    "setuptools<81" \
+    https://github.com/csukuangfj/piper-phonemize/releases/download/2025.06.23/piper_phonemize-1.3.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl \
+    git+https://github.com/ysharma3501/LinaCodec.git \
+    git+https://github.com/ysharma3501/LuxTTS.git
 
-COPY ./wheel /home/user/app/wheel
 COPY ./requirements.txt /home/user/app/requirements.txt
-WORKDIR /home/user/app
 RUN pip install --no-cache-dir -r /home/user/app/requirements.txt
 
 # LuxTTS voice cloning engine
@@ -100,11 +115,14 @@ ENV PATH="/home/user/app/venv/bin:$PATH" \
     LD_LIBRARY_PATH="/usr/local/cuda/lib64:${LD_LIBRARY_PATH}"
 
 COPY --chown=1001:1001 --from=builder /home/user/app/venv /home/user/app/venv
+
+
+FROM runtime AS app
 COPY ./modules /home/user/app/modules
 COPY ./wheel /home/user/app/wheel
 COPY ./tests /home/user/app/tests
 COPY ./docs /home/user/app/docs
-COPY ./config.json /home/user/app/config.json
+# COPY ./config.json /home/user/app/config.json
 COPY ./voice_clone_studio.py /home/user/app/voice_clone_studio.py
 WORKDIR /home/user/app
 EXPOSE 7860

README.md

@@ -1,6 +1,6 @@
 # Voice Clone Studio
 
-A modular Gradio-based web UI for voice cloning, voice design, and multi-speaker conversation, powered by [Qwen3-TTS](https://github.com/QwenLM/Qwen3-TTS), [VibeVoice](https://github.com/microsoft/VibeVoice) and [LuxTTS](https://github.com/ysharma3501/LuxTTS)
+A modular Gradio-based web UI for voice cloning, voice design, and multi-speaker conversation, powered by [Qwen3-TTS](https://github.com/QwenLM/Qwen3-TTS), [VibeVoice](https://github.com/microsoft/VibeVoice), and [LuxTTS](https://github.com/ysharma3501/LuxTTS).
 Supports both Whisper and VibeVoice ASR for automatic transcription.
 
 ![Voice Clone Studio](https://img.shields.io/badge/Voice%20Clone%20Studio-v1.0-blue) ![Qwen3-TTS](https://img.shields.io/badge/Qwen3--TTS-Powered-blue) ![LuxTTS](https://img.shields.io/badge/LuxTTS-TTS-orange) ![VibeVoice](https://img.shields.io/badge/VibeVoice-TTS-green) ![VibeVoice](https://img.shields.io/badge/VibeVoice-ASR-green) 
@@ -14,7 +14,7 @@ Voice Clone Studio is fully modular. The main file dynamically loads self-contai
 ### Voice Clone
 Clone voices from your own audio samples. Provide a short reference audio clip with its transcript, and generate new speech in that voice.
 
-- **Multiple engines** - Qwen3-TTS (0.6B/1.7B) or VibeVoice (1.5B/Large/Large-4bit)
+- **Multiple engines** - Qwen3-TTS (0.6B/1.7B), VibeVoice (1.5B/Large/Large-4bit), or LuxTTS (Default)
 - **Voice prompt caching** - First generation processes the sample, subsequent ones are instant
 - **Seed control** - Reproducible results with saved seeds
 - **Emotion presets** - 40+ emotion presets with adjustable intensity

docker-compose.yaml

@@ -2,21 +2,25 @@ services:
   voice-clone-studio:
     build:
       context: .
-      target: runtime
+      target: app
     ports:
       - "7860:7860"
+    env_file: .env
     environment:
       - "GRADIO_SERVER_NAME=0.0.0.0"
       - "HF_HOME=${HF_CACHE_CONTAINER_PATH:-/home/user/app/.cache/huggingface}"
+      - "TORCH_HOME=${TORCH_CACHE_CONTAINER_PATH:-/home/user/app/.cache/torch}"
       - "NVIDIA_VISIBLE_DEVICES=all"
       - "NVIDIA_DRIVER_CAPABILITIES=compute,utility"
     volumes:
+      - "./config.json:/home/user/app/config.json"
       - "./temp:/home/user/app/temp"
       - "./datasets:/home/user/app/datasets"
       - "./models:/home/user/app/models"
       - "./samples:/home/user/app/samples"
       - "./output:/home/user/app/output"
       - "${HF_CACHE_HOST_PATH:-./.hf_cache}:${HF_CACHE_CONTAINER_PATH:-/home/user/app/.cache/huggingface}"
+      - "${TORCH_CACHE_HOST_PATH:-./.torch_cache}:${TORCH_CACHE_CONTAINER_PATH:-/home/user/app/.cache/torch}"
     deploy:
       resources:
         reservations:

docs/updates.md

@@ -11,6 +11,7 @@
 ## February 7, 2026
 
 #### Version 1.0.0 - Complete Modular Rewrite
+- **LuxTTS Voice Clone Engine** - Added LuxTTS as a third voice cloning engine with 48 kHz output, caching, and UI controls
 - **Full Modular Architecture** - Complete rewrite from a 6000+ line monolith into independent tool modules under `modules/core_components/tools/`
 - **Tool System** - Each tab is now a self-contained tool with its own UI, events, and logic, loaded dynamically from a central registry
 - **Enable/Disable Tools** - New "Visible Tools" section in Settings lets you toggle any tab on or off (persisted in config, takes effect on restart)

modules/core_components/__init__.py

@@ -8,7 +8,7 @@
     INPUT_MODAL_CSS,
     INPUT_MODAL_HEAD,
     INPUT_MODAL_HTML,
-    show_input_modal_js
+    show_input_modal_js,
 )
 
 from .emotion_manager import (
@@ -17,7 +17,7 @@
     get_emotion_choices,
     calculate_emotion_values,
     handle_save_emotion,
-    handle_delete_emotion
+    handle_delete_emotion,
 )
 
 from .constants import (
@@ -35,6 +35,7 @@
     DEFAULT_CONFIG,
     QWEN_GENERATION_DEFAULTS,
     VIBEVOICE_GENERATION_DEFAULTS,
+    LUXTTS_DEFAULTS,
 )
 
 __all__ = [
@@ -70,4 +71,5 @@
     "DEFAULT_CONFIG",
     "QWEN_GENERATION_DEFAULTS",
     "VIBEVOICE_GENERATION_DEFAULTS",
+    "LUXTTS_DEFAULTS",
 ]