A dry-stone yard has ASR transcripts and intent tags but no spoken-back lines for the kiosk. What does that mean for the conversational pipeline corpus?
Select an answer to reveal the explanation.
Short Explanation
Hearing and understanding without a scripted answer leaves the talker silent. The gap is on the TTS speak arm—write reply lines, don’t redesign fusion to dodge them.
Full Explanation
A conversational pipeline needs a speak arm: spoken-style reply scripts for TTS. Transcripts and intents alone leave that modality missing. Redesigning fusion or swapping in wall photos does not create the reply corpus.