Overview
Transcripts provide a structured record of meeting conversations.
Recommended: Use sentence transcripts (/sentenced-transcripts/*) for the best developer experience (speaker labels, ordering, timestamps).
Need recordings instead?
Recording endpoints are documented separately.
Get All Sentence Transcripts for a Scenario
Retrieve all sentence transcripts for meetings conducted with a specific scenario.
curl -X GET https://api.waterrai.com/sentenced-transcripts/scenario/456 \
-H "Authorization: Bearer YOUR_API_SECRET_OR_JWT_TOKEN"
{
"success": true,
"data": {
"scenario_info": {
"id": 456,
"name": "Product Manager Interview Practice",
"description": "Practice interview scenarios"
},
"meetings": [
{
"meeting_id": 789,
"person_name": "John Doe",
"created_at": "2024-01-15T10:40:00Z",
"transcripts": [
{
"transcript_id": 301,
"sentences": [
{
"id": 1001,
"speaker": "user",
"message": "Hello, I'm excited to practice this interview.",
"start_duration": "00:00",
"sequence_number": 1
}
]
}
]
}
]
}
}
Get Sentence Transcripts for a Meeting
Retrieve sentence transcripts for a specific meeting.
curl -X GET https://api.waterrai.com/sentenced-transcripts/301/789/456 \
-H "Authorization: Bearer YOUR_API_SECRET_OR_JWT_TOKEN"
{
"success": true,
"data": {
"sentences": [
{
"id": 1001,
"transcript_id": 301,
"speaker": "user",
"message": "Hello, I'm excited to practice this interview.",
"start_duration": "00:00",
"sequence_number": 1
}
],
"meeting_info": {
"id": 789
},
"scenario_info": {
"id": 456,
"name": "Product Manager Interview Practice",
"description": "Practice interview scenarios"
}
}
}
Get Transcripts by Meeting ID (non-sentenced)
Retrieve all transcripts (non-sentenced) for a specific meeting.
curl -X GET https://api.waterrai.com/transcripts/meeting/789 \
-H "Authorization: Bearer YOUR_API_SECRET_OR_JWT_TOKEN"
[
{
"id": 301,
"meeting_id": 789,
"content": "Full transcript text...",
"status": "completed",
"language": "en",
"created_at": "2024-01-15T11:10:00Z"
}
]
If you’re building product features on top of transcripts, prefer sentence transcripts because they’re already normalized for parsing and ordering.