Durable Batch Audio/Video Transcription
MeteredTool ID: helve.audio_transcribe
Submit an audio or video URL for durable batch transcription via AssemblyAI, Deepgram, or ElevenLabs Scribe. Returns a job ID to poll at GET /v1/jobs/{id}.
Input Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
audio_url | string | Required | HTTPS URL of the recording, fetched by the provider. Signed URLs are fine; it must stay reachable until the provider has read it. No embedded credentials or fragments. |
provider | string (assemblyai | deepgram | elevenlabs | auto) | Optional | Which speech-to-text provider runs the job. `auto` (default) picks the first configured provider. The choice is persisted on the job before execution. |
language | string | Optional | BCP-47-style language tag such as `en` or `en-US`. Omit to let the provider detect the language. |
diarize | boolean | Optional | Label speakers in `words` as `speaker_0`, `speaker_1`, … in order of first appearance. Default false. |
keyterms | array | Optional | Recognition hints: names, jargon, product terms. Supported on AssemblyAI, Deepgram Nova-3, and ElevenLabs Scribe v2; other models ignore them with a warning. |
timestamps | string (word | none) | Optional | `word` (default) returns per-word timings in `result.words`; `none` omits them. |
strict | boolean | Optional | When true, any option the selected provider or model cannot honour rejects the request instead of producing a warning. Default false. |
provider_options | object | Optional | Provider-native options, validated by the adapter. Model selection lives here. Transport settings such as URLs, callbacks, and authentication cannot be overridden. |
Call via REST API
curl -X POST https://helve.dev/v1/tools/call \
-H "Authorization: Bearer sk_live_YOUR_KEY" \
-H "Content-Type: application/json" \
-d '{
"tool_id": "helve.audio_transcribe",
"arguments": {
"audio_url": "https://example.com/meeting.mp3",
"diarize": true
}
}'
Call via MCP (call_tool)
{
"name": "call_tool",
"arguments": {
"tool_id": "helve.audio_transcribe",
"arguments": {
"audio_url": "https://example.com/meeting.mp3",
"diarize": true
}
}
}