Merge pull request #15 from kadirnar/feature/whisplus-enhancements

feat: 🚀 Add GitHub workflows for package testing and CI, integrate logger support in whisper.py 🤖

.github/FUNDING.yml

@@ -0,0 +1 @@
+github: kadirnar

.github/workflows/ci.yml

@@ -0,0 +1,79 @@
+name: Python package CI
+
+on:
+  push:
+    branches: [main]
+  pull_request:
+    branches: [main]
+
+jobs:
+  build:
+    runs-on: ${{ matrix.operating-system }}
+    strategy:
+      matrix:
+        operating-system: [ubuntu-latest, windows-latest, macos-latest]
+        python-version: [3.8, 3.9, 3.10, 3.11]
+        torch-version:  [2.1.0, 2.1.1]
+      fail-fast: false
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v2
+
+      - name: Set up Python
+        uses: actions/setup-python@v2
+        with:
+          python-version: ${{ matrix.python-version }}
+
+      - name: Restore Ubuntu cache
+        uses: actions/cache@v2
+        if: matrix.operating-system == 'ubuntu-latest'
+        with:
+          path: ~/.cache/pip
+          key: ${{ matrix.operating-system }}-${{ matrix.python-version }}-${{ hashFiles('**/setup.py')}}
+          restore-keys: ${{ matrix.operating-system }}-${{ matrix.python-version }}-
+
+      - name: Restore MacOS cache
+        uses: actions/cache@v2
+        if: matrix.operating-system == 'macos-latest'
+        with:
+          path: ~/Library/Caches/pip
+          key: ${{ matrix.operating-system }}-${{ matrix.python-version }}-${{ hashFiles('**/setup.py')}}
+          restore-keys: ${{ matrix.operating-system }}-${{ matrix.python-version }}-
+
+      - name: Restore Windows cache
+        uses: actions/cache@v2
+        if: matrix.operating-system == 'windows-latest'
+        with:
+          path: ~\AppData\Local\pip\Cache
+          key: ${{ matrix.operating-system }}-${{ matrix.python-version }}-${{ hashFiles('**/setup.py')}}
+          restore-keys: ${{ matrix.operating-system }}-${{ matrix.python-version }}-
+
+      - name: Update pip
+        run: python -m pip install --upgrade pip
+
+      - name: Check styling with pre-commit
+        run: |
+          pre-commit install
+          pre-commit run --all-files
+
+      - name: Install PyTorch on Linux and Windows
+        if: >
+          matrix.operating-system == 'ubuntu-latest' ||
+          matrix.operating-system == 'windows-latest'
+        run: >
+          pip install torch==${{ matrix.torch-version }}+cpu
+          -f https://download.pytorch.org/whl/torch_stable.html
+
+      - name: Install PyTorch on MacOS
+        if: matrix.operating-system == 'macos-latest'
+        run: pip install torch==${{ matrix.torch-version }}
+      - name: Install PyTorch on Linux and Windows
+
+      - name: Install whisperplus package from local setup.py
+        run: >
+          pip install -e .
+
+      - name: Unittest whisperplus
+        run: |
+          python -m unittest

.github/workflows/package_testing.yml

@@ -0,0 +1,78 @@
+name: Package Testing
+
+on:
+  push:
+    branches: [ master ]
+  pull_request:
+    branches: [ master ]
+
+jobs:
+  build:
+    runs-on: ${{ matrix.operating-system }}
+
+    strategy:
+      matrix:
+        operating-system: [ubuntu-latest, windows-latest, macos-latest]
+        python-version: [3.8, 3.9, 3.10, 3.11]
+        torch-version: [2.1.0, 2.1.1]
+      fail-fast: false
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v2
+
+      - name: Set up Python
+        uses: actions/setup-python@v2
+        with:
+          python-version: ${{ matrix.python-version }}
+
+      - name: Restore Ubuntu cache
+        uses: actions/cache@v2
+        if: matrix.operating-system == 'ubuntu-latest'
+        with:
+          path: ~/.cache/pip
+          key: ${{ matrix.operating-system }}-${{ matrix.python-version }}-${{ hashFiles('**/setup.py')}}
+          restore-keys: ${{ matrix.operating-system }}-${{ matrix.python-version }}-
+
+      - name: Restore MacOS cache
+        uses: actions/cache@v2
+        if: matrix.operating-system == 'macos-latest'
+        with:
+          path: ~/Library/Caches/pip
+          key: ${{ matrix.operating-system }}-${{ matrix.python-version }}-${{ hashFiles('**/setup.py')}}
+          restore-keys: ${{ matrix.operating-system }}-${{ matrix.python-version }}-
+
+      - name: Restore Windows cache
+        uses: actions/cache@v2
+        if: matrix.operating-system == 'windows-latest'
+        with:
+          path: ~\AppData\Local\pip\Cache
+          key: ${{ matrix.operating-system }}-${{ matrix.python-version }}-${{ hashFiles('**/setup.py')}}
+          restore-keys: ${{ matrix.operating-system }}-${{ matrix.python-version }}-
+
+      - name: Update pip
+        run: python -m pip install --upgrade pip
+
+      - name: Install Numpy
+        run: >
+          pip install numpy
+
+      - name: Install PyTorch on Linux and Windows
+        if: >
+          matrix.operating-system == 'ubuntu-latest' ||
+          matrix.operating-system == 'windows-latest'
+        run: >
+          pip install torch==${{ matrix.torch-version }}+cpu
+          -f https://download.pytorch.org/whl/torch_stable.html
+
+      - name: Install PyTorch on MacOS
+        if: matrix.operating-system == 'macos-latest'
+        run: pip install torch==${{ matrix.torch-version }}
+
+      - name: Install latest whisperplus package
+        run: >
+          pip install --upgrade --force-reinstall whisperplus
+
+      - name: Unittest whisperplus
+        run: |
+          python -m unittest

.github/workflows/publish_pypi.yml

@@ -0,0 +1,34 @@
+name: Publish Python distributions to PyPI
+
+on:
+  push:
+    branches: [ master ]
+
+jobs:
+  build-n-publish:
+    name: Build and publish Python distributions to PyPI
+    runs-on: ubuntu-latest
+
+    steps:
+    - uses: actions/checkout@v2
+    - name: Set up Python 3.8
+      uses: actions/setup-python@v2
+      with:
+        python-version: 3.8
+    - name: Install pypa/build
+      run: >-
+        python -m
+        pip install
+        build
+    - name: Build a binary wheel and a source tarball
+      run: >-
+        python -m
+        build
+        --sdist
+        --wheel
+        --outdir dist/
+    - name: Publish distribution 📦 to PyPI
+      uses: pypa/gh-action-pypi-publish@master
+      with:
+        user: __token__
+        password: ${{ secrets.PYPI_WHISPERPLUS_API_KEY  }}

whisperplus/__init__.py

@@ -2,7 +2,7 @@
 from whisperplus.pipelines.whisper import SpeechToTextPipeline
 from whisperplus.utils.download_utils import download_and_convert_to_mp3
 
-__version__ = '0.0.4'
+__version__ = '0.0.5'
 __author__ = 'kadirnar'
 __license__ = 'Apache License 2.0'
 __all__ = ['']

whisperplus/pipelines/whisper.py

@@ -1,6 +1,10 @@
+import logging
+
 import torch
 from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor, pipeline
 
+logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
+
 
 class SpeechToTextPipeline:
     """Class for converting audio to text using a pre-trained speech recognition model."""
@@ -11,6 +15,8 @@ def __init__(self, model_id: str = "openai/whisper-large-v3"):
 
         if self.model is None:
             self.load_model(model_id)
+        else:
+            logging.info("Model already loaded.")
 
         self.set_device()
 
@@ -21,16 +27,20 @@ def set_device(self):
         else:
             self.device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 
+        logging.info(f"Using device: {self.device}")
+
     def load_model(self, model_id: str = "openai/whisper-large-v3"):
         """
         Loads the pre-trained speech recognition model and moves it to the specified device.
 
         Args:
             model_id (str): Identifier of the pre-trained model to be loaded.
         """
+        logging.info("Loading model...")
         model = AutoModelForSpeechSeq2Seq.from_pretrained(
             model_id, torch_dtype=torch.float16, low_cpu_mem_usage=True, use_safetensors=True)
         model.to(self.device)
+        logging.info("Model loaded successfully.")
 
         self.model = model
 
@@ -60,6 +70,6 @@ def __call__(self, audio_path: str, model_id: str = "openai/whisper-large-v3", l
             model_kwargs={"use_flash_attention_2": True},
             generate_kwargs={"language": language},
         )
-
+        logging.info("Transcribing audio...")
         result = pipe(audio_path)["text"]
         return result