Alignment Timestamps
Extract word-level timestamps by aligning a transcript to an audio file.
POST
/api/v1/alignment/extract-timestampsRequest Body
| Parameter | Type | Required | Description |
|---|---|---|---|
| model | string | Yes | Model key (see models below) |
| params | object | Yes | Model-specific parameters |
| attachmentMediaIds | string[] | Yes | Audio file to align against |
Response
{
"inferenceJobId": "ij_..."
}Example
POST
/api/v1/alignment/extract-timestamps{
"model": "elevenlabs-extract-timestamps",
"params": {
"text": "Welcome to Wonda, the AI-powered video editing platform."
},
"attachmentMediaIds": ["media_abc123"]
}Models
elevenlabs-extract-timestamps
Align a transcript to audio and extract word-level timestamps via ElevenLabs.
| Parameter | Type | Required | Description |
|---|---|---|---|
| text | string | Yes | The transcript text to align against the audio |
Attachments: 1 required audio file (wav, mp3, mpeg, m4a).