Upload 10 files

.gitignore

@@ -0,0 +1,31 @@
+# Ignore Python bytecode
+__pycache__
+
+# Ignore settings
+config.json
+.env
+
+# Ignore generated files
+outputs/
+favorites/
+voices/
+collections/
+outputs-rvc/
+voices-tortoise/
+
+# Ignore model checkpoints
+data/
+
+# Ignore temporary files
+temp/
+
+# Ignore node modules
+node_modules/
+
+# Editors
+.vscode/
+.idea/
+.sourcery.yaml
+
+# Ignore temporary Jupiter notebooks
+*.temp.ipynb

Dockerfile

@@ -0,0 +1,33 @@
+# Python 3.10 w/ Nvidia Cuda
+FROM nvidia/cuda:11.8.0-devel-ubuntu22.04 AS env_base
+
+# Install Pre-reqs
+RUN apt-get update && apt-get install --no-install-recommends -y \
+    git vim nano build-essential python3-dev python3-venv python3-pip gcc g++ ffmpeg
+
+# Setup venv
+RUN pip3 install virtualenv
+RUN virtualenv /venv
+ENV VIRTUAL_ENV=/venv
+RUN python3 -m venv $VIRTUAL_ENV
+ENV PATH="$VIRTUAL_ENV/bin:$PATH"
+RUN pip3 install --upgrade pip setuptools && \
+    pip3 install torch torchvision torchaudio
+
+# Set working directory
+WORKDIR /app
+
+# Clone the repo
+RUN git clone https://github.com/rsxdalv/tts-generation-webui.git
+
+# Set working directory to the cloned repo
+WORKDIR /app/tts-generation-webui
+
+# Install all requirements
+RUN pip3 install -r requirements.txt
+RUN pip3 install -r requirements_audiocraft.txt
+RUN pip3 install -r requirements_bark_hubert_quantizer.txt
+RUN pip3 install -r requirements_rvc.txt
+
+# Run the server
+CMD python server.py

LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2023 Roberts Slisans
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

app.py

@@ -0,0 +1,160 @@
+import os
+import src.utils.setup_or_recover as setup_or_recover
+import src.utils.dotenv_init as dotenv_init
+import gradio as gr
+
+from src.config.load_config import default_config
+from src.config.config import config
+
+from src.css.css import full_css
+from src.Joutai import Joutai
+from src.history_tab.collections_directories_atom import collections_directories_atom
+
+setup_or_recover.dummy()
+dotenv_init.init()
+
+
+def reload_config_and_restart_ui():
+    os._exit(0)
+    # print("Reloading config and restarting UI...")
+    # config = load_config()
+    # gradio_interface_options = config["gradio_interface_options"] if "gradio_interface_options" in config else {}
+    # demo.close()
+    # time.sleep(1)
+    # demo.launch(**gradio_interface_options)
+
+
+gradio_interface_options = (
+    config["gradio_interface_options"]
+    if "gradio_interface_options" in config
+    else default_config
+)
+
+with gr.Blocks(
+    css=full_css,
+    title="TTS Generation WebUI",
+    analytics_enabled=False,  # it broke too many times
+) as demo:
+    gr.Markdown("# TTS Generation WebUI (Bark, MusicGen + AudioGen, Tortoise, RVC)")
+    with Joutai.singleton.tabs:
+        from src.tortoise.generation_tab_tortoise import generation_tab_tortoise
+        from src.settings_tab_gradio import settings_tab_gradio
+        from src.bark.generation_tab_bark import generation_tab_bark
+        from src.history_tab.main import history_tab
+        from src.bark.settings_tab_bark import settings_tab_bark
+        from src.history_tab.voices_tab import voices_tab
+        from src.vocos.vocos_tabs import vocos_tabs
+        from src.studio.studio_tab import simple_remixer_tab
+
+        register_use_as_history_button = generation_tab_bark()
+
+        try:
+            from src.bark.clone.tab_voice_clone import tab_voice_clone
+
+            tab_voice_clone(register_use_as_history_button)
+        except Exception as e:
+            from src.bark.clone.tab_voice_clone_error import tab_voice_clone_error
+
+            tab_voice_clone_error(e)
+            print("Failed to load voice clone demo")
+            print(e)
+
+        try:
+            from src.musicgen.musicgen_tab import generation_tab_musicgen
+
+            generation_tab_musicgen()
+        except Exception as e:
+            from src.musicgen.musicgen_tab_error import musicgen_tab_error
+
+            musicgen_tab_error(e)
+            print("Failed to load musicgen demo")
+            print(e)
+
+        try:
+            from src.rvc_tab.rvc_tab import rvc_conversion_tab
+
+            rvc_conversion_tab()
+        except Exception as e:
+            from src.rvc_tab.rvc_tab_error import rvc_tab_error
+
+            rvc_tab_error(e)
+            print("Failed to load rvc demo")
+            print(e)
+
+        try:
+            from src.demucs.demucs_tab import demucs_tab
+
+            demucs_tab()
+        except Exception as e:
+            from src.demucs.demucs_tab_error import demucs_tab_error
+
+            demucs_tab_error(e)
+            print("Failed to load demucs demo")
+            print(e)
+
+        vocos_tabs()
+        generation_tab_tortoise()
+
+        collections_directories_atom.render()
+        history_tab(register_use_as_history_button)
+        history_tab(register_use_as_history_button, directory="favorites")
+        history_tab(
+            register_use_as_history_button, directory="outputs", show_collections=True
+        )
+        voices_tab(register_use_as_history_button)
+
+        with gr.Tab("Settings"):
+            from src.settings_tab_gradio import settings_tab_gradio
+
+            settings_tab_gradio(reload_config_and_restart_ui, gradio_interface_options)
+
+            from src.bark.settings_tab_bark import settings_tab_bark
+
+            settings_tab_bark()
+            from src.utils.model_location_settings_tab import (
+                model_location_settings_tab,
+            )
+
+            model_location_settings_tab()
+
+        remixer_input = simple_remixer_tab()
+    Joutai.singleton.tabs.render()
+
+
+def print_pretty_options(options):
+    print("Gradio interface options:")
+    max_key_length = max(len(key) for key in options.keys())
+    for key, value in options.items():
+        if key == "auth" and value is not None:
+            print(f"  {key}:{' ' * (max_key_length - len(key))} {value[0]}:******")
+        else:
+            print(f"  {key}:{' ' * (max_key_length - len(key))} {value}")
+
+
+# detect if --share is passed
+if "--share" in os.sys.argv:
+    print("Gradio share mode enabled")
+    gradio_interface_options["share"] = True
+
+
+print("Starting Gradio server...")
+if not gradio_interface_options["enable_queue"]:
+    print("Warning: Gradio server queue is disabled. Automatically enabling")
+    gradio_interface_options["enable_queue"] = True
+if gradio_interface_options["auth"] is not None:
+    # split username:password into (username, password)
+    gradio_interface_options["auth"] = tuple(
+        gradio_interface_options["auth"].split(":")
+    )
+    print("Gradio server authentication enabled")
+print_pretty_options(gradio_interface_options)
+
+
+def start_server():
+    demo.queue(
+        concurrency_count=gradio_interface_options.get("concurrency_count", 5),
+    ).launch(**gradio_interface_options)
+
+
+if __name__ == "__main__":
+    start_server()

docker-compose.yml

@@ -0,0 +1,15 @@
+version: '3'
+services:
+  tts-generation-webui:
+    image: rsxdalv/tts-generation-webui
+    restart: unless-stopped
+    ports:
+      - "7860:7860"
+    container_name: tts-generation-webui
+    deploy:
+        resources:
+          reservations:
+            devices:
+              - driver: nvidia
+                device_ids: ['0']
+                capabilities: [gpu]

requirements.txt

@@ -0,0 +1,10 @@
+safetensors==0.3.1 # Until they fix it
+torch==2.0.0 # BSD 3-Clause - ensures torch is not reinstalled
+ffmpeg-python # Apache 2.0
+gradio==3.35.2
+python-dotenv==1.0.0
+soundfile==0.12.1; sys_platform == 'win32' # torchaudio
+# sox # torchaudio for linux
+suno-bark @ git+https://github.com/suno-ai/bark@56b0ba13f7c281cbffa07ea9abf7b30273a60b6a#egg=suno-bark # MIT License
+vocos==0.0.2 # MIT License
+tortoise @ git+https://github.com/rsxdalv/tortoise-tts@72eccabcb7d4c7a9d943a2c9b63211e4286cf385#egg=tortoise # Apache 2.0

requirements_audiocraft.txt

@@ -0,0 +1,4 @@
+torch==2.0.0 # BSD 3-Clause - ensures torch is not reinstalled
+# audiocraft==0.0.1 # MIT License
+xformers==0.0.19 # For torch==2.0.0 project plane
+audiocraft @ git+https://[email protected]/facebookresearch/audiocraft@e96018613ac82b1afe0f0cce7861dfe08ba2b3bf#egg=audiocraft # MIT License

requirements_bark_hubert_quantizer.txt

@@ -0,0 +1,4 @@
+torch==2.0.0 # BSD 3-Clause - ensures torch is not reinstalled
+# pre-install fairseq for bark_hubert_quantizer
+fairseq @ https://github.com/Sharrnah/fairseq/releases/download/v0.12.4/fairseq-0.12.4-cp310-cp310-win_amd64.whl ; sys_platform == 'win32' # MIT License
+bark_hubert_quantizer @ git+https://github.com/rsxdalv/bark-voice-cloning-HuBERT-quantizer@bark_hubert_quantizer#egg=bark_hubert_quantizer # MIT License

requirements_rvc.txt

@@ -0,0 +1,3 @@
+torch==2.0.0 # BSD 3-Clause - ensures torch is not reinstalled
+torchcrepe @ git+https://github.com/rsxdalv/torchcrepe@patch-1 # MIT License
+rvc-beta @ git+https://github.com/rsxdalv/Retrieval-based-Voice-Conversion-WebUI@package # MIT License

update.py

@@ -0,0 +1,33 @@
+import subprocess
+
+
+def do(cmd):  # https://stackoverflow.com/a/62986640
+    try:
+        run = subprocess.run(cmd, shell=True)
+        run.check_returncode()
+        return run
+    except subprocess.CalledProcessError as e:
+        print(e.stderr.decode().strip())
+        raise e
+
+
+def try_install(requirements, name=None):
+    try:
+        print(f"Installing {name or requirements} dependencies...")
+        do(f"pip install -r {requirements}")
+        print(f"Successfully installed {name or requirements} dependencies")
+    except Exception:
+        print(f"Failed to install {name or requirements} dependencies")
+
+
+def main():
+    print("Updating dependencies...")
+    try_install("requirements_audiocraft.txt", "musicgen, audiocraft")
+    try_install("requirements_bark_hubert_quantizer.txt", "Bark Voice Clone, bark-hubert-quantizer")
+    try_install("requirements_rvc.txt", "RVC")
+    # hydracore fix because of fairseq
+    do("pip install hydra-core==1.3.2")
+
+
+if __name__ == "__main__":
+    main()