Fix installer with uv and swift f0 pitcher

.github/workflows/build-container-image.yml

@@ -19,7 +19,7 @@ jobs:
       id-token: write
     steps:
       - name: Checkout
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
       - name: Inject enhanced GitHub environment variables
         uses: rlespinasse/github-slug-action@v5 # https://github.com/rlespinasse/github-slug-action
       - name: lowercase IMAGE_REGISTRY

.github/workflows/main.yml

@@ -3,13 +3,13 @@ name: Check Requirements
 on:
   push:
     paths:
-      - "requirements-linux.txt"
-      - "requirements-windows.txt"
+      - "pyproject.toml"
+      - "uv.lock"
 
   pull_request:
     paths:
-      - "requirements-linux.txt"
-      - "requirements-windows.txt"
+      - "pyproject.toml"
+      - "uv.lock"
 
 jobs:
   build:
@@ -23,29 +23,35 @@ jobs:
             TARGET: macOS
           - os: windows-latest
             TARGET: Windows
-        python-version: [ "3.10" ]
+        python-version: [ "3.12" ]
 
     steps:
     - name: Checkout repository
-      uses: actions/checkout@v3
+      uses: actions/checkout@v6
 
-    - name: Set up Python 3.10
-      uses: actions/setup-python@v4
+    - name: Set up Python 3.12
+      uses: actions/setup-python@v6
       with:
         python-version: ${{ matrix.python-version }}
 
-    - name: Install dependencies (Windows)
-      if: matrix.TARGET == 'Windows'
+    - name: Install uv
       run: |
-        python -m pip install --upgrade pip
-        pip install -r requirements-windows.txt
+        curl -LsSf https://astral.sh/uv/install.sh | sh
+      shell: bash
 
-    - name: Install dependencies (others)
+    - name: Add uv to PATH (Unix)
       if: matrix.TARGET != 'Windows'
-      run: |
-        python -m pip install --upgrade pip
-        pip install -r requirements-linux.txt
-
-    - name: Install gpu dependencies
-      if: matrix.TARGET != 'macOS'
-      run: pip3 install torch==2.0.1+cu117 torchvision==0.15.2+cu117 torchaudio==2.0.2+cu117 --index-url https://download.pytorch.org/whl/cu117
+      run: echo "$HOME/.local/bin" >> $GITHUB_PATH
+      shell: bash
+
+    - name: Add uv to PATH (Windows)
+      if: matrix.TARGET == 'Windows'
+      run: echo "$env:USERPROFILE\.local\bin" | Out-File -FilePath $env:GITHUB_PATH -Encoding utf8 -Append
+      shell: pwsh
+
+    - name: Sync dependencies from pyproject.toml
+      run: uv sync
+      shell: bash
+
+
+

.github/workflows/pylint.yml

@@ -14,10 +14,10 @@ jobs:
         python-version: ["3.10"]
     steps:
     - name: Checkout code
-      uses: actions/checkout@v3
+      uses: actions/checkout@v6
 
     - name: Set up Python ${{ matrix.python-version }}
-      uses: actions/setup-python@v3
+      uses: actions/setup-python@v6
       with:
         python-version: ${{ matrix.python-version }}
 

.github/workflows/pytest.yml

@@ -7,30 +7,25 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       matrix:
-        python-version: [ "3.10" ]
+        python-version: [ "3.12" ]
 
     steps:
     - name: Checkout code
-      uses: actions/checkout@v3
+      uses: actions/checkout@v6
 
     - name: Set up Python ${{ matrix.python-version }}
-      uses: actions/setup-python@v3
+      uses: actions/setup-python@v6
       with:
         python-version: ${{ matrix.python-version }}
 
-    - name: Install dependencies (Windows)
-      if: matrix.TARGET == 'Windows'
+    - name: Install uv
       run: |
-        python -m pip install --upgrade pip
-        pip install -r requirements-windows.txt
-        #pip install -e .
+        curl -LsSf https://astral.sh/uv/install.sh | sh
+        echo "$HOME/.local/bin" >> $GITHUB_PATH
 
-    - name: Install dependencies (others)
-      if: matrix.TARGET != 'Windows'
-      run: |
-        python -m pip install --upgrade pip
-        pip install -r requirements-linux.txt
-        #pip install -e .
+    - name: Sync dependencies from pyproject.toml
+      run: uv sync
 
     - name: Run pytest
-      run: python -m pytest --import-mode=append pytest/
+      run: uv run pytest --import-mode=append pytest/
+

Dockerfile

@@ -1,30 +1,44 @@
-FROM nvidia/cuda:12.6.3-runtime-ubuntu22.04
+FROM nvidia/cuda:12.8.1-runtime-ubuntu22.04
 
-# note: the python3-pip package contains Python 3.10 on Ubuntu 22.04
+# Set timezone and configure non-interactive installation
+ENV DEBIAN_FRONTEND=noninteractive
+ENV TZ=UTC
+
+# Install Python 3.12 from deadsnakes PPA and build tools
 RUN apt-get update \
-    && apt-get install git python3-pip python3.10-venv ffmpeg -y  \
+    && apt-get install -y software-properties-common \
+    && add-apt-repository ppa:deadsnakes/ppa -y \
+    && apt-get update \
+    && apt-get install -y git python3.12 python3.12-venv python3.12-dev ffmpeg curl tzdata \
+       build-essential gcc g++ make \
     && apt-get clean  \
     && rm -rf /var/lib/apt/lists/*
 
-# copy only the requirements file to leverage container image build cache
-COPY ./requirements-linux.txt /app/UltraSinger/requirements-linux.txt
-WORKDIR /app/UltraSinger
+# Install uv
+RUN curl -LsSf https://astral.sh/uv/install.sh | sh
+ENV PATH="/root/.local/bin:$PATH"
+ENV UV_LINK_MODE=copy
 
-# no need to run as root
-RUN chown -R 1000:1000 /app/UltraSinger
-USER 1000:1000
+# copy pyproject.toml first to leverage container image build cache
+COPY ./pyproject.toml /app/UltraSinger/pyproject.toml
+# Need to copy some minimal source structure for editable install
+RUN mkdir -p /app/UltraSinger/src
+WORKDIR /app/UltraSinger
 
-# setup venv
-ENV VIRTUAL_ENV=/app/UltraSinger/.venv
-RUN python3 -m venv $VIRTUAL_ENV
-ENV PATH="$VIRTUAL_ENV/bin:$PATH"
+# Install dependencies from pyproject.toml directly without venv (container is already isolated)
+# Using build isolation (without --no-build-isolation) so uv handles all build dependencies automatically
+RUN uv pip install --system --python 3.12 -e .
 
-# install dependencies
-RUN pip install --no-cache-dir -r requirements-linux.txt \
-    && pip install --no-cache-dir torch==2.3.0 torchvision==0.18.0 torchaudio==2.3.0 --index-url https://download.pytorch.org/whl/cu121 \
-    && pip install --no-cache-dir tensorflow[and-cuda]==2.16.1
+# Install PyTorch with CUDA support (override the CPU version from pyproject.toml)
+RUN uv pip install --system --python 3.12 torch==2.8.0 torchvision==0.23.0 torchaudio==2.8.0 --index-url https://download.pytorch.org/whl/cu128 --reinstall
 
 # copy sources late to allow for caching of layers which contain all the dependencies
 COPY . /app/UltraSinger
+
+
+# no need to run as root
+RUN chown -R 1000:1000 /app/UltraSinger
+USER 1000:1000
+
 WORKDIR /app/UltraSinger/src
-CMD ["bash" ]
+CMD ["bash"]

README.md

@@ -69,16 +69,16 @@ This will help me a lot to keep this project alive and improve it.
 
 ### Installation
 
-* Install Python 3.10 **(older and newer versions has some breaking changes)**. [Download](https://www.python.org/downloads/)
+* Install Python 3.12 **(older/newer versions may have compatibility issues)**. [Download](https://www.python.org/downloads/)
 * Also download or install ffmpeg with PATH. [Download](https://www.ffmpeg.org/download.html)
-* Go to folder `install` and run install script for your OS.
-  * Choose `GPU` if you have an nvidia CUDA GPU.
-  * Choose `CPU` if you don't have an nvidia CUDA GPU.
+* Go to folder `install` and run install script for your OS:
+  * Choose `GPU` if you have an NVIDIA CUDA GPU.
+  * Choose `CPU` if you don't have an NVIDIA GPU or want CPU-only processing.
 
 ### Run
 
-* In root folder just run `run_on_windows.bat`, `run_on_linux.sh` or `run_on_macos.command` to start the app.
-* Now you can use the UltraSinger source code with `py UltraSinger.py [opt] [mode] [transcription] [pitcher] [extra]`. See [How to use](#how-to-use) for more information.
+* In root folder just run `run_on_windows.bat`, `run_on_linux.sh` or `run_on_mac.command` to start the app.
+* Now you can use the UltraSinger source code with `py UltraSinger.py [opt] [mode] [transcription] [pitcher] [extra]`. See [How to use](#-how-to-use-the-app) for more information.
 
 ## 📖 How to use the App
 
@@ -119,11 +119,6 @@ _Not all options working now!_
     --whisper_compute_type  Change to "int8" if low on GPU mem (may reduce accuracy) >> ((default) is "float16" for cuda devices, "int8" for cpu)
     --keep_numbers          Numbers will be transcribed as numerics instead of as words
 
-    [pitcher]
-    # Default is crepe
-    --crepe            tiny|full >> ((default) is full)
-    --crepe_step_size  unit is miliseconds >> ((default) is 10)
-
     [extra]
     --disable_hyphenation   Disable word hyphenation. Hyphenation is enabled by default.
     --disable_separation    Disable track separation. Track separation is enabled by default.
@@ -143,7 +138,6 @@ _Not all options working now!_
     [device]
     --force_cpu             Force all steps to be processed on CPU.
     --force_whisper_cpu     Only whisper will be forced to cpu
-    --force_crepe_cpu       Only crepe will be forced to cpu
 ```
 
 For standard use, you only need to use [opt]. All other options are optional.
@@ -219,14 +213,9 @@ starts at the place or is heard. To disable:
 
 ### 👂 Pitcher
 
-Pitching is done with the `crepe` model.
-Also consider that a bigger model is more accurate, but also takes longer to pitch.
-For just testing you should use `tiny`.
-If you want solid accurate, then use the `full` model.
-
-```commandline
--i XYZ --crepe full
-```
+Pitching is done with the `SwiftF0` model, which is faster and more accurate than CREPE.
+SwiftF0 automatically detects pitch frequencies between 46.875 Hz (G1) and 2093.75 Hz (C7).
+UltraSinger uses 60hz and 400hz
 
 ### 👄 Separation
 
@@ -285,33 +274,24 @@ this MIDI and sheet are created. And you also want to have accurate files
 
 With a GPU you can speed up the process. Also the quality of the transcription and pitching is better.
 
-You need a cuda device for this to work. Sorry, there is no cuda device for macOS.
-
-It is optional (but recommended) to install the cuda driver for your gpu: see [driver](https://developer.nvidia.com/cuda-downloads).
-Install torch with cuda separately in your `venv`. See [tourch+cuda](https://pytorch.org/get-started/locally/).
-Also check you GPU cuda support. See [cuda support](https://gist.github.com/standaloneSA/99788f30466516dbcc00338b36ad5acf)
+You need an NVIDIA CUDA device for this to work. Sorry, there is no CUDA device for macOS.
 
-Command for `pip`:
-```
-pip3 install torch==2.0.1+cu117 torchvision==0.15.2+cu117 torchaudio==2.0.2+cu117 --index-url https://download.pytorch.org/whl/cu117
-```
-
-When you want to use `conda` instead you need a [different installation command](https://pytorch.org/get-started/locally/).
+For GPU support on Windows and Linux, the installation script automatically installs PyTorch with CUDA support.
 
-#### Considerations for Windows users
+It is optional (but recommended) to install the CUDA driver for your GPU: see [CUDA driver](https://developer.nvidia.com/cuda-downloads).
+Also check your GPU CUDA support. See [CUDA support](https://gist.github.com/standaloneSA/99788f30466516dbcc00338b36ad5acf)
 
-The pitch tracker used by UltraSinger (crepe) uses TensorFlow as its backend.
-TensorFlow dropped GPU support for Windows for versions >2.10 as you can see in this [release note](https://github.com/tensorflow/tensorflow/releases/tag/v2.11.1) and their [installation instructions](https://www.tensorflow.org/install/pip#windows-native).
-
-For now UltraSinger runs the latest version available that still supports GPUs on windows.
+For manual installation, you can use:
+```bash
+uv pip install --index-url https://download.pytorch.org/whl/cu121 torch torchvision torchaudio
+```
 
-For running later versions of TensorFlow on windows while still taking advantage of GPU support the suggested solution is to [run UltraSinger in a container](container/README.md).
 #### Crashes due to low VRAM
 
-If something crashes because of low VRAM then use a smaller model.
+If something crashes because of low VRAM then use a smaller Whisper model.
 Whisper needs more than 8GB VRAM in the `large` model!
 
-You can also force cpu usage with the extra option `--force_cpu`.
+You can also force CPU usage with the extra option `--force_cpu`.
 
 ### 📦 Containerized (Docker or Podman)
 

ReleaseNotes.md

@@ -14,6 +14,9 @@ Date: 2026.02.10
   - Support for video as input
   - Optimise scale detection
   - Added quantization by key
+  - Changed installer to uv
+  - Drop crepe for SwiftF0
+  - upgrade to python 3.12
 
 # Version: 0.0.12
 Date: 2024.12.19

colab/UltraSinger.ipynb

@@ -25,19 +25,35 @@
    "outputs": [],
    "source": [
     "%cd /content\n",
+    "\n",
+    "# Remove existing directory if present\n",
+    "#!rm -rf UltraSinger\n",
+    "\n",
+    "# Clone specific branch\n",
     "!git clone https://github.com/rakuri255/UltraSinger.git\n",
     "%cd /content/UltraSinger\n",
-    "!pip install --no-cache-dir -r requirements-linux.txt\n",
-    "!pip install --no-cache-dir torch==2.3.0 torchvision==0.18.0 torchaudio==2.3.0 --index-url https://download.pytorch.org/whl/cu121\n",
-    "!pip install --no-cache-dir tensorflow[and-cuda]==2.17.1\n",
+    "\n",
+    "# Install uv\n",
+    "!curl -LsSf https://astral.sh/uv/install.sh | sh\n",
+    "\n",
+    "# Add uv to PATH and configure matplotlib\n",
+    "import os\n",
+    "os.environ['PATH'] = f\"/root/.local/bin:{os.environ['PATH']}\"\n",
+    "os.environ['UV_LINK_MODE'] = 'copy'\n",
+    "os.environ['MPLBACKEND'] = 'Agg'\n",
+    "\n",
+    "# Sync dependencies from pyproject.toml\n",
+    "!uv sync\n",
+    "\n",
+    "# Install PyTorch with CUDA support\n",
+    "!uv pip install torch==2.8.0 torchvision==0.23.0 torchaudio==2.8.0 --index-url https://download.pytorch.org/whl/cu128 --force-reinstall\n",
+    "\n",
     "%cd /content/UltraSinger/src"
    ]
   },
   {
    "cell_type": "code",
-   "source": [
-    "!python UltraSinger.py -i https://www.youtube.com/watch?v=YwNs1Z0qRY0 -o /content/output"
-   ],
+   "source": "!../.venv/bin/python UltraSinger.py -i https://www.youtube.com/watch?v=YwNs1Z0qRY0 -o /content/output\n",
    "metadata": {
     "id": "O0j4vUW0YAG2"
    },

install/CPU/linux_cpu.sh

@@ -1,7 +1,35 @@
 #!/bin/bash
-cd ..
-cd ..
-python3.10 -m venv .venv
-source .venv/bin/activate
-pip install -r requirements-linux.txt
-pip install torch==2.3.1 torchvision==0.18.1 torchaudio==2.3.1
+set -e
+
+cd "$(dirname "$0")"
+cd ../..
+
+# Set link mode to copy to avoid hardlink warnings
+export UV_LINK_MODE=copy
+
+# Install uv if not already installed
+if ! command -v uv &> /dev/null; then
+    echo "Installing uv..."
+    curl -LsSf https://astral.sh/uv/install.sh | sh
+    # Update PATH for current session
+    export PATH="$HOME/.local/bin:$PATH"
+fi
+
+# Verify uv is available
+if ! command -v uv &> /dev/null; then
+    echo "Error: uv could not be found or installed"
+    echo "Please ensure your shell PATH includes ~/.local/bin"
+    exit 1
+fi
+
+echo "uv version:"
+uv --version
+
+echo "Syncing dependencies with uv..."
+uv sync --extra linux
+
+echo "Installation completed successfully!"
+echo "To run UltraSinger:"
+echo "  source .venv/bin/activate"
+echo "  cd src"
+echo "  py UltraSinger.py"