Last released Oct 23, 2026
Speech-to-text that notices what you're talking about and biases Whisper toward the right names.