{
"message_type": "conversation",
"event_type": "conversation.perception_tool_call",
"seq": 129,
"conversation_id": "your-conversation-id",
"turn_idx": 3,
"properties": {
"name": "describe_scene",
"arguments": "{\"detail\":\"high\"}",
"modality": "audio",
"frames": [
{
"uri": "https://example.invalid/frame"
}
]
}
}Perception Tool Call Event
Outbound: Tavus broadcasts when Raven triggers a perception tool from visual or audio input.
Configure perception tools: Perception tool calling and Perception.
conversation for conversation-layer app messages.
"conversation"
conversation.perception_tool_call.
"conversation.perception_tool_call"
Unix timestamp (seconds since epoch) indicating when this event was created. Use this to build timestamped transcripts or reconstruct the timeline of a conversation.
1746000000.123
Monotonic per event; a higher seq was emitted later. Use with turn_idx as in Event ordering and turn tracking.
42
The unique identifier for the conversation.
"c123456"
Turn index; increments when conversation.respond is processed. Groups utterances, tool calls, and speaking-state events for the same turn. See Event ordering and turn tracking.
3
Tool name, modality from the perception step, arguments as returned by the model, optional non-empty frames for multimodal metadata.
Show child attributes
Show child attributes

