A gilder’s loft has leaf-and-frame photographs and no spoken turns. A contractor calls the album “the multimodal Riva set.” How should that claim be handled?
Select an answer to reveal the explanation.
Short Explanation
Pretty pictures of gold leaf are not a conversation. Riva’s pipeline needs spoken turns and text—hear, understand, speak—not a stills drawer.
Full Explanation
Conversational Riva data is built from speech and text turns, including reply scripts when TTS is in scope. A stills-only album is not that corpus. Kernel work and fusion-architecture labels belong to other domains and do not rescue the claim.