Get transcript
Full transcript of a meeting with speaker and time per line.
MCP tool tldv_read → get_transcript
read
Returns the transcript of one meeting, one line per speaker turn with its timestamp. While tl;dv is still processing the recording, the call returns status processing and no text, so try again later. Use the timestamps with create_clip and add_note.
Parameters
| Name | Type | Description |
|---|---|---|
| meeting_id string required | string | Meeting ID from list_meetings. |
tools/call
{
"jsonrpc": "2.0",
"id": 1,
"method": "tools/call",
"params": {
"name": "tldv_read",
"arguments": {
"operation": "get_transcript",
"arguments": {
"meeting_id": "meeting_8c41d2"
}
}
}
}Example response
{
"content": [
{
"type": "text",
"text": "{ ... same JSON as structuredContent ... }"
}
],
"structuredContent": {
"meeting_id": "mt_4f21a9",
"status": "ready",
"lines": [
{
"time": "00:12:05",
"speaker": "Dana Whitfield",
"text": "We have the budget confirmed for Q4, but we need SSO before launch."
},
{
"time": "00:12:21",
"speaker": "You",
"text": "Understood, SSO is on the plan for November."
}
]
},
"isError": false
}