speaker-encoder-pipeline 0.1.0__tar.gz

speaker_encoder_pipeline-0.1.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2025 Speaker Encoder Pipeline Contributors
+
+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.

speaker_encoder_pipeline-0.1.0/MANIFEST.in

@@ -0,0 +1,19 @@
+include README.md
+include LICENSE
+include version.txt
+include requirements.txt
+include requirements-dev.txt
+include pytest.ini
+
+recursive-include model *.py
+recursive-include data *.py
+recursive-include train *.py
+recursive-include utils *.py
+recursive-include tests *.py
+
+recursive-exclude * __pycache__
+recursive-exclude * *.pyc
+recursive-exclude * *.pyo
+recursive-exclude * .DS_Store
+recursive-exclude wandb *
+recursive-exclude models *

speaker_encoder_pipeline-0.1.0/PKG-INFO

@@ -0,0 +1,321 @@
+Metadata-Version: 2.4
+Name: speaker-encoder-pipeline
+Version: 0.1.0
+Summary: Multilingual Speaker Encoder with Multi-branch Architecture for Few-shot Voice Synthesis
+Home-page: https://github.com/yourusername/Zero_shotVoiceClone
+Author: Your Name
+Author-email: Your Name <your.email@example.com>
+License: MIT
+Project-URL: Homepage, https://github.com/yourusername/Zero_shotVoiceClone
+Project-URL: Bug Reports, https://github.com/yourusername/Zero_shotVoiceClone/issues
+Project-URL: Source Code, https://github.com/yourusername/Zero_shotVoiceClone
+Keywords: speaker-encoder,voice-synthesis,multilingual,few-shot,deep-learning
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Classifier: Topic :: Multimedia :: Sound/Audio :: Speech
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: torch>=2.0.0
+Requires-Dist: torchaudio>=2.0.1
+Requires-Dist: librosa>=0.9.0
+Requires-Dist: numpy>=1.21.0
+Requires-Dist: pandas>=1.3.0
+Requires-Dist: scikit-learn>=1.0.0
+Requires-Dist: tqdm>=4.60.0
+Requires-Dist: wandb>=0.13.0
+Provides-Extra: dev
+Requires-Dist: pytest>=7.0.0; extra == "dev"
+Requires-Dist: pytest-cov>=4.0.0; extra == "dev"
+Requires-Dist: pytest-xdist>=3.0.0; extra == "dev"
+Requires-Dist: black>=22.0.0; extra == "dev"
+Requires-Dist: flake8>=4.0.0; extra == "dev"
+Requires-Dist: isort>=5.0.0; extra == "dev"
+Requires-Dist: mypy>=0.990; extra == "dev"
+Requires-Dist: build>=0.10.0; extra == "dev"
+Requires-Dist: twine>=4.0.0; extra == "dev"
+Provides-Extra: test
+Requires-Dist: pytest>=7.0.0; extra == "test"
+Requires-Dist: pytest-cov>=4.0.0; extra == "test"
+Requires-Dist: pytest-xdist>=3.0.0; extra == "test"
+Dynamic: author
+Dynamic: home-page
+Dynamic: license-file
+Dynamic: requires-python
+
+# Speaker Encoder Pipeline
+
+A multilingual speaker encoder with multi-branch architecture for few-shot voice synthesis. This package provides a robust speaker encoding system that captures timbre, cadence, and self-supervised learning features.
+
+## Features
+
+- **Multi-branch Architecture**: Combines timbre, cadence, and SSL features
+- **Multilingual Support**: English and Tamil language support
+- **Few-shot Learning**: Efficient speaker encoding with minimal audio samples
+- **Modular Design**: Easy to extend and customize
+- **CI/CD Ready**: Fully configured with GitHub Actions
+
+## Installation
+
+### From PyPI (when published)
+
+```bash
+pip install speaker-encoder-pipeline
+```
+
+### From Source
+
+```bash
+git clone https://github.com/yourusername/Zero_shotVoiceClone.git
+cd speaker_encoder_pipeline
+
+# Install in development mode
+pip install -e ".[dev]"
+```
+
+## Quick Start
+
+### Basic Usage
+
+```python
+from model.encoder import MultilingualSpeakerEncoder
+from config import SAMPLE_RATE
+import torch
+
+# Initialize model
+encoder = MultilingualSpeakerEncoder(config)
+
+# Prepare audio (mono, 16kHz)
+audio = torch.randn(batch_size, SAMPLE_RATE * duration)
+
+# Get speaker embedding
+embedding = encoder(audio)
+print(embedding.shape)  # (batch_size, embedding_dim)
+```
+
+### Command-line Interface
+
+```bash
+# Prepare data
+speaker-encoder-prepare --data-dir ./data
+
+# Train model
+speaker-encoder-train --epochs 100 --batch-size 32
+
+# Validate speakers
+speaker-encoder-validate --model-path ./models/best_model.pt
+
+# Evaluate performance
+speaker-encoder-evaluate --test-dir ./test_data
+```
+
+## Development Setup
+
+### Installation with Dev Dependencies
+
+```bash
+pip install -e ".[dev]"
+```
+
+### Running Tests
+
+```bash
+# Run all tests
+pytest
+
+# Run with verbose output
+pytest -v
+
+# Run specific test file
+pytest tests/test_model.py
+
+# Run specific test
+pytest tests/test_model.py::TestMultilingualSpeakerEncoder::test_forward_pass
+
+# Run with coverage report
+pytest --cov=model --cov=data --cov=train --cov=utils --cov-report=html
+
+# Run only unit tests
+pytest -m unit
+
+# Run without slow tests
+pytest -m "not slow"
+
+# Run in parallel (faster)
+pytest -n auto
+```
+
+### Code Quality
+
+```bash
+# Format code with black
+black model/ data/ train/ utils/
+
+# Check imports with isort
+isort model/ data/ train/ utils/
+
+# Lint with flake8
+flake8 model/ data/ train/ utils/
+
+# Type checking with mypy
+mypy model/ data/ train/ utils/
+```
+
+## Package Structure
+
+```
+speaker_encoder_pipeline/
+├── model/
+│   ├── encoder.py          # Main encoder architecture
+│   ├── timbre_branch.py    # Timbre extraction
+│   ├── cadence_branch.py   # Cadence extraction
+│   ├── ssl_branch.py       # SSL features
+│   ├── fusion.py           # Feature fusion
+│   └── loss.py             # Loss functions
+├── data/
+│   ├── dataset.py          # Dataset classes
+│   ├── preprocessor.py     # Audio preprocessing
+│   ├── augmentations.py    # Data augmentation
+│   ├── sampler.py          # Sampling strategies
+│   └── __init__.py
+├── train/
+│   ├── dataloader.py       # Data loading
+│   └── __init__.py
+├── utils/
+│   ├── audio_utils.py      # Audio utilities
+│   └── __init__.py
+├── tests/
+│   ├── test_model.py       # Model tests
+│   ├── test_data.py        # Data tests
+│   ├── test_train.py       # Training tests
+│   ├── test_package.py     # Package structure tests
+│   └── conftest.py         # Pytest configuration
+├── config.py               # Configuration
+├── setup.py                # Package setup
+├── pyproject.toml          # Modern Python packaging
+├── pytest.ini              # Pytest configuration
+├── requirements.txt        # Core dependencies
+├── requirements-dev.txt    # Development dependencies
+└── README.md               # This file
+```
+
+## Configuration
+
+Edit `config.py` to customize:
+
+- Audio parameters (sample rate, channels, format)
+- Dataset paths and sources
+- Model architecture parameters
+- Training hyperparameters
+
+## Publishing to PyPI
+
+### 1. Update Version
+
+```bash
+# Update version in version.txt
+echo "0.2.0" > version.txt
+
+# Commit changes
+git add version.txt
+git commit -m "Bump version to 0.2.0"
+```
+
+### 2. Create a Git Tag
+
+```bash
+git tag -a v0.2.0 -m "Release version 0.2.0"
+git push origin v0.2.0
+```
+
+### 3. GitHub Actions will automatically:
+- Run all tests on multiple Python versions and OSes
+- Build distribution packages (wheel and sdist)
+- Publish to PyPI (requires `PYPI_API_TOKEN` secret in GitHub)
+- Create a GitHub Release
+
+### 4. Configure PyPI Token (one-time setup)
+
+In your GitHub repository:
+1. Go to **Settings → Secrets and variables → Actions**
+2. Add a new secret: `PYPI_API_TOKEN`
+3. Generate token at https://pypi.org/manage/account/token/
+
+## Testing
+
+The project uses pytest with the following markers:
+
+- `@pytest.mark.unit` - Unit tests
+- `@pytest.mark.integration` - Integration tests
+- `@pytest.mark.slow` - Slow tests
+- `@pytest.mark.gpu` - GPU-required tests
+- `@pytest.mark.model` - Model-specific tests
+- `@pytest.mark.data` - Data processing tests
+- `@pytest.mark.train` - Training tests
+
+## CI/CD Workflow
+
+The GitHub Actions workflow (`python-package.yml`):
+
+1. **Test Job**: Runs on every push and pull request
+   - Tests on Ubuntu, Windows, and macOS
+   - Python 3.9, 3.10, 3.11
+   - Linting with flake8
+   - Code formatting with black
+   - Import sorting with isort
+   - Coverage reports to Codecov
+
+2. **Build and Publish Job**: Runs on version tags
+   - Builds distribution packages
+   - Validates with twine
+   - Publishes to PyPI
+   - Creates GitHub Release
+
+## Contributing
+
+1. Fork the repository
+2. Create a feature branch: `git checkout -b feature/your-feature`
+3. Make changes and write tests
+4. Ensure all tests pass: `pytest`
+5. Format code: `black` and `isort`
+6. Commit and push to your fork
+7. Create a Pull Request
+
+## License
+
+MIT License - see LICENSE file for details
+
+## Citation
+
+If you use this package in your research, please cite:
+
+```bibtex
+@software{speaker_encoder_2025,
+  title={Speaker Encoder Pipeline},
+  author={Your Name},
+  year={2025},
+  url={https://github.com/yourusername/Zero_shotVoiceClone}
+}
+```
+
+## Support
+
+For issues and questions:
+- Open an issue on GitHub
+- Check existing documentation
+- Review test cases for usage examples
+
+## Changelog
+
+### Version 0.1.0 (Initial Release)
+- Multi-branch speaker encoder
+- Multilingual support (English, Tamil)
+- Complete training pipeline
+- CI/CD with GitHub Actions
+- PyPI packaging support

speaker_encoder_pipeline-0.1.0/README.md

@@ -0,0 +1,270 @@
+# Speaker Encoder Pipeline
+
+A multilingual speaker encoder with multi-branch architecture for few-shot voice synthesis. This package provides a robust speaker encoding system that captures timbre, cadence, and self-supervised learning features.
+
+## Features
+
+- **Multi-branch Architecture**: Combines timbre, cadence, and SSL features
+- **Multilingual Support**: English and Tamil language support
+- **Few-shot Learning**: Efficient speaker encoding with minimal audio samples
+- **Modular Design**: Easy to extend and customize
+- **CI/CD Ready**: Fully configured with GitHub Actions
+
+## Installation
+
+### From PyPI (when published)
+
+```bash
+pip install speaker-encoder-pipeline
+```
+
+### From Source
+
+```bash
+git clone https://github.com/yourusername/Zero_shotVoiceClone.git
+cd speaker_encoder_pipeline
+
+# Install in development mode
+pip install -e ".[dev]"
+```
+
+## Quick Start
+
+### Basic Usage
+
+```python
+from model.encoder import MultilingualSpeakerEncoder
+from config import SAMPLE_RATE
+import torch
+
+# Initialize model
+encoder = MultilingualSpeakerEncoder(config)
+
+# Prepare audio (mono, 16kHz)
+audio = torch.randn(batch_size, SAMPLE_RATE * duration)
+
+# Get speaker embedding
+embedding = encoder(audio)
+print(embedding.shape)  # (batch_size, embedding_dim)
+```
+
+### Command-line Interface
+
+```bash
+# Prepare data
+speaker-encoder-prepare --data-dir ./data
+
+# Train model
+speaker-encoder-train --epochs 100 --batch-size 32
+
+# Validate speakers
+speaker-encoder-validate --model-path ./models/best_model.pt
+
+# Evaluate performance
+speaker-encoder-evaluate --test-dir ./test_data
+```
+
+## Development Setup
+
+### Installation with Dev Dependencies
+
+```bash
+pip install -e ".[dev]"
+```
+
+### Running Tests
+
+```bash
+# Run all tests
+pytest
+
+# Run with verbose output
+pytest -v
+
+# Run specific test file
+pytest tests/test_model.py
+
+# Run specific test
+pytest tests/test_model.py::TestMultilingualSpeakerEncoder::test_forward_pass
+
+# Run with coverage report
+pytest --cov=model --cov=data --cov=train --cov=utils --cov-report=html
+
+# Run only unit tests
+pytest -m unit
+
+# Run without slow tests
+pytest -m "not slow"
+
+# Run in parallel (faster)
+pytest -n auto
+```
+
+### Code Quality
+
+```bash
+# Format code with black
+black model/ data/ train/ utils/
+
+# Check imports with isort
+isort model/ data/ train/ utils/
+
+# Lint with flake8
+flake8 model/ data/ train/ utils/
+
+# Type checking with mypy
+mypy model/ data/ train/ utils/
+```
+
+## Package Structure
+
+```
+speaker_encoder_pipeline/
+├── model/
+│   ├── encoder.py          # Main encoder architecture
+│   ├── timbre_branch.py    # Timbre extraction
+│   ├── cadence_branch.py   # Cadence extraction
+│   ├── ssl_branch.py       # SSL features
+│   ├── fusion.py           # Feature fusion
+│   └── loss.py             # Loss functions
+├── data/
+│   ├── dataset.py          # Dataset classes
+│   ├── preprocessor.py     # Audio preprocessing
+│   ├── augmentations.py    # Data augmentation
+│   ├── sampler.py          # Sampling strategies
+│   └── __init__.py
+├── train/
+│   ├── dataloader.py       # Data loading
+│   └── __init__.py
+├── utils/
+│   ├── audio_utils.py      # Audio utilities
+│   └── __init__.py
+├── tests/
+│   ├── test_model.py       # Model tests
+│   ├── test_data.py        # Data tests
+│   ├── test_train.py       # Training tests
+│   ├── test_package.py     # Package structure tests
+│   └── conftest.py         # Pytest configuration
+├── config.py               # Configuration
+├── setup.py                # Package setup
+├── pyproject.toml          # Modern Python packaging
+├── pytest.ini              # Pytest configuration
+├── requirements.txt        # Core dependencies
+├── requirements-dev.txt    # Development dependencies
+└── README.md               # This file
+```
+
+## Configuration
+
+Edit `config.py` to customize:
+
+- Audio parameters (sample rate, channels, format)
+- Dataset paths and sources
+- Model architecture parameters
+- Training hyperparameters
+
+## Publishing to PyPI
+
+### 1. Update Version
+
+```bash
+# Update version in version.txt
+echo "0.2.0" > version.txt
+
+# Commit changes
+git add version.txt
+git commit -m "Bump version to 0.2.0"
+```
+
+### 2. Create a Git Tag
+
+```bash
+git tag -a v0.2.0 -m "Release version 0.2.0"
+git push origin v0.2.0
+```
+
+### 3. GitHub Actions will automatically:
+- Run all tests on multiple Python versions and OSes
+- Build distribution packages (wheel and sdist)
+- Publish to PyPI (requires `PYPI_API_TOKEN` secret in GitHub)
+- Create a GitHub Release
+
+### 4. Configure PyPI Token (one-time setup)
+
+In your GitHub repository:
+1. Go to **Settings → Secrets and variables → Actions**
+2. Add a new secret: `PYPI_API_TOKEN`
+3. Generate token at https://pypi.org/manage/account/token/
+
+## Testing
+
+The project uses pytest with the following markers:
+
+- `@pytest.mark.unit` - Unit tests
+- `@pytest.mark.integration` - Integration tests
+- `@pytest.mark.slow` - Slow tests
+- `@pytest.mark.gpu` - GPU-required tests
+- `@pytest.mark.model` - Model-specific tests
+- `@pytest.mark.data` - Data processing tests
+- `@pytest.mark.train` - Training tests
+
+## CI/CD Workflow
+
+The GitHub Actions workflow (`python-package.yml`):
+
+1. **Test Job**: Runs on every push and pull request
+   - Tests on Ubuntu, Windows, and macOS
+   - Python 3.9, 3.10, 3.11
+   - Linting with flake8
+   - Code formatting with black
+   - Import sorting with isort
+   - Coverage reports to Codecov
+
+2. **Build and Publish Job**: Runs on version tags
+   - Builds distribution packages
+   - Validates with twine
+   - Publishes to PyPI
+   - Creates GitHub Release
+
+## Contributing
+
+1. Fork the repository
+2. Create a feature branch: `git checkout -b feature/your-feature`
+3. Make changes and write tests
+4. Ensure all tests pass: `pytest`
+5. Format code: `black` and `isort`
+6. Commit and push to your fork
+7. Create a Pull Request
+
+## License
+
+MIT License - see LICENSE file for details
+
+## Citation
+
+If you use this package in your research, please cite:
+
+```bibtex
+@software{speaker_encoder_2025,
+  title={Speaker Encoder Pipeline},
+  author={Your Name},
+  year={2025},
+  url={https://github.com/yourusername/Zero_shotVoiceClone}
+}
+```
+
+## Support
+
+For issues and questions:
+- Open an issue on GitHub
+- Check existing documentation
+- Review test cases for usage examples
+
+## Changelog
+
+### Version 0.1.0 (Initial Release)
+- Multi-branch speaker encoder
+- Multilingual support (English, Tamil)
+- Complete training pipeline
+- CI/CD with GitHub Actions
+- PyPI packaging support

speaker_encoder_pipeline-0.1.0/data/augmentations.py

@@ -0,0 +1,67 @@
+import torch
+import torchaudio
+import random
+import numpy as np
+
+class AudioAugmentations:
+    """On-the-fly audio augmentations for speaker encoder training."""
+    
+    def __init__(self, config):
+        self.config = config
+        self.noise_snr_min = config.NOISE_SNR_MIN
+        self.noise_snr_max = config.NOISE_SNR_MAX
+        self.reverb_prob = config.REVERB_PROB
+        self.speed_prob = config.SPEED_PERTURB_PROB
+        self.speed_factors = config.SPEED_FACTORS
+    
+    def __call__(self, audio):
+        """Apply augmentations to audio tensor."""
+        audio = audio.clone()
+        
+        # Speed perturbation
+        if random.random() < self.speed_prob:
+            factor = random.choice(self.speed_factors)
+            audio = self._speed_perturb(audio, factor)
+        
+        # Add noise
+        if self.noise_snr_min > 0:
+            audio = self._add_noise(audio)
+        
+        # Reverb (simplified - can be enhanced with actual RIRs)
+        if random.random() < self.reverb_prob:
+            audio = self._add_reverb(audio)
+        
+        return audio
+    
+    def _speed_perturb(self, audio, factor):
+        """Change playback speed."""
+        # Simple resampling
+        indices = torch.arange(0, len(audio), factor)
+        indices = indices[indices < len(audio)].long()
+        return audio[indices]
+    
+    def _add_noise(self, audio):
+        """Add random noise with specified SNR."""
+        # Generate random noise
+        noise = torch.randn_like(audio) * 0.01
+        
+        # Calculate target SNR
+        snr_db = random.uniform(self.noise_snr_min, self.noise_snr_max)
+        snr = 10 ** (snr_db / 20)
+        
+        # Scale noise to achieve target SNR
+        audio_power = torch.mean(audio ** 2)
+        noise_power = torch.mean(noise ** 2)
+        noise = noise * torch.sqrt(audio_power / (noise_power * snr ** 2))
+        
+        return audio + noise
+    
+    def _add_reverb(self, audio):
+        """Simple reverb simulation (impulse response placeholder)."""
+        # Simple delay effect
+        delay_len = 160  # 10ms at 16kHz
+        if len(audio) > delay_len:
+            delayed = torch.zeros_like(audio)
+            delayed[delay_len:] = audio[:-delay_len] * 0.3
+            return audio + delayed
+        return audio