Focused crawls are collections of frequently-updated webcrawl data from narrow (as opposed to broad or wide) web crawls, often focused on a single domain or subdomain.
A multimodal corpus that can be used to train English-to-Italian End-to-End Speech-to-Text Machine Translation models, English ASR models, or English-to-Italian/Italian-to-English statistical textual Machine Translation models