# Speech Verification System using X-Vector Architecture
A robust speaker verification system implemented in Python using PyTorch. The system uses X-Vector architecture with MFCC and pitch features for speaker verification tasks.
## Features
- X-Vector deep neural network architecture for speaker embedding
- Combined MFCC and pitch features for improved verification
- K-fold cross-validation for robust model evaluation
- Early stopping mechanism to prevent overfitting
- Automatic data balancing for positive and negative samples
- Comprehensive training visualization and model checkpoint saving
## Requirements
```
See environment.yml
```
## Project Structure
```
├── dataset/ # Dataset directory
├── speaker1/ # Target speaker directory
└── speaker2/ # Other speakers directory
....
└── speakern/ # Other n speakers directory
├── environment.yml …
