ktalk_get_transcript
Retrieve speaker-labeled transcripts from KTalk recordings. Supports raw JSON or markdown with timecodes, and paged chunks for reading large recordings in manageable parts.
Instructions
Get transcript of a KTalk recording (speech-to-text by speakers).
Args: recording_key: Recording key/identifier (required) format: Output format — "raw" (JSON) or "markdown" (dialogue with timecodes) chunk: Chunk number. 0 = auto (returns full text if small, first chunk if large). 1+ = specific chunk number for paged reading. chunk_size: Max characters per chunk (~7500 tokens at 30000). Soft limit — chunks split at utterance boundaries, never mid-utterance.
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"recording_key": "<recording_key>"
},
"name": "ktalk_get_transcript"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| chunk | No | ||
| format | No | markdown | |
| chunk_size | No | ||
| recording_key | Yes |
Output Schema
| Name | Required | Description | Default |
|---|---|---|---|
| result | Yes |