Skip to main content
POST
GetVoiceWidgetConversationAPI mirrors erdo_voice_widget_conversation_get,

Authorizations

Authorization
string
header
required

An Erdo API key (erdo_api_...) or scoped token (erdo_token_...).

Body

application/json
session_id
string
widget
string

Response

Success response

contact
Contact is what a person told us about themselves. The four fields are the · object

ones the leads dataset has columns for; there is deliberately no company, because nothing downstream would use it.

created_at
string<date-time>
duration_seconds
integer<int32>
Required range: -2147483648 <= x <= 2147483647
ended_at
string<date-time>
has_transcript
boolean
lead_saved_at
string<date-time>
lead_status
string

still awaiting one, then created | existing | disabled | test.

modality
string
origin
string
session_id
string
topics
string[] | null

analysed yet", [] is "analysed, nothing found".

transcript_json
TranscriptJSON is the raw stored transcript (ElevenLabs post-call array or our · object

own turns array), passed through verbatim. Null when no transcript was captured.

transcript_summary
string
transcript_text
string
transcript_turns
WidgetTranscriptTurn is one turn of a conversation as a consumer displays it: · object[]

displaying a conversation renders ONE shape instead of re-parsing both the ElevenLabs post-call array and our own turns array. Built by the same tolerant renderer as TranscriptText, so the two can never disagree. Always present (empty when nothing rendered).

turns
WidgetConversationTurn is one agent turn's LLM performance metrics, extracted · object[]

from an ElevenLabs transcript (empty for our simpler text/video turn shapes, which carry no LLM metrics). See extractWidgetConversationTurns.

widget
string
widget_id
string
widget_name
string