# TPU Training

Hubert (Hidden-Unit BERT) is a speech representation learning model proposed by Facebook, which uses self-supervised learning to directly learn speech features from raw waveform signals.

This is a Stable Diffusion model trained on TPU and converted to PyTorch, specializing in text-to-image generation tasks, particularly for plush toy image generation.

SanBERTa is a RoBERTa model trained on Sanskrit text, specifically designed for Sanskrit text processing tasks.