Using pretrained encoder and language models to generate captions from multimedia inputs. - View it on GitHub