TeluguGPT is built using a transformer-based architecture, specifically fine-tuned for English to Telugu translation. The model has been trained on a large parallel corpus of English-Telugu text, ensuring high accuracy and contextual understanding. The project also includes extensive pre-processing techniques like tokenization and normalization to handle the nuances of the Telugu language, making it highly effective for real-world translation tasks.
- Transformer-based architecture for high-quality and contextually accurate translations.
- English to Telugu Translation capability.
- Customizable for domain-specific fine-tuning and training.
- Python 3
- Required Python libraries:
transformerstorchnumpytokenizersdatasets
- Clone the repository:
git clone https://github.com/uddithmachiraju/TeluguGPT.git