Business Problem
:
When an AI Agent is assigned to a conversation, it does not process audio transcripts that were previously triggered by a human or replies by AI Assist. As a result, those transcribed audio messages and replies are excluded from the AI's context, even when they are within the last 20 messages.
This creates context gap
:
• AI Agents may respond without full conversation history
• Important customer intent captured in voice notes is ignored
• AI Agents produce inconsistent or incorrect responses due to missing information
Desired Outcome
:
When an AI Agent is assigned, it can understand past voice interactions via transcripts.
Use Cases
:
• Handling conversations where customers send context through audio messages
• Enabling AI Agents to use manually transcribed voice notes when generating replies
• Preserving conversation context when ownership shifts from human agents to AI Agents