Gemini 1.5
Google DeepMind — Gemini
rich detail 9 method fields published
Evaluation
CoVoST-2 (speech translation; Gemini 1.5)
Speech translation benchmark: 20 languages into English, subset seen during pre-training; BLEU metric
How it was run
Open a release to see the setup, scoring, and other details that the lab published. Only fields the lab actually disclosed are shown.