This repository provides a PyTorch implementation of the Conformer architecture from the INTERSPEECH 2020 paper. Conformer combines the global modeling capability of transformers with the local feature extraction of convolutional neural networks to process audio sequences. The model achieves state-of-the-art results in speech recognition tasks by efficiently capturing both local and global dependencies in speech signals. This is a model-focused implementation; training capabilities are available through the associated openspeech project.