Transcribe an audio or video file to text

Run job:audio.transcribe through the looot API: 1 provider, from $0.000072 per result. Inputs, prices, output shape and code for curl, JavaScript and Python.

Transcribe an audio or video file to text.

Job id audio.transcribe, in Generative AI (Media generation). As of 2026-10-01, 1 provider serve it through 1 tool. Run job:audio.transcribe and looot picks one of them; with fallback on, a miss moves on to the next. See Jobs.

Inputs

This job has one tool, so it takes that tool’s fields. Any key you send passes through unchanged.

Name	Type	Required
`url`	string	yes
`callback`	string	no
`callback_method`	string	no
`custom_intent`	string	no
`custom_intent_mode`	string	no
`custom_topic`	string	no
`custom_topic_mode`	string	no
`detect_entities`	boolean	no
`detect_language`	string	no
`diarize`	boolean	no
`diarize_model`	string	no
`dictation`	boolean	no
`encoding`	string	no
`extra`	string	no
`filler_words`	boolean	no
`intents`	boolean	no
`keyterm`	array	no
`keywords`	string	no
`language`	string	no
`measurements`	boolean	no
`mip_opt_out`	boolean	no
`model`	string	no
`multichannel`	boolean	no
`numerals`	boolean	no
`paragraphs`	boolean	no
`profanity_filter`	boolean	no
`punctuate`	boolean	no
`redact`	string	no
`replace`	string	no
`search`	string	no
`sentiment`	boolean	no
`smart_format`	boolean	no
`summarize`	string	no
`tag`	string	no
`topics`	boolean	no
`utt_split`	number	no
`utterances`	boolean	no
`version`	string	no

Providers and prices

Provider	Tool	Price
Deepgram	`deepgram-listen`	$0.000072 per result

A call that fails at the provider costs $0. See What’s free.

Output

Shape of the run’s result, checked against 3 real answers:

{ results: { channels: { alternatives: { ... }[] }[] }, metadata: { models: string[], sha256, created, channels, duration, model_info: Record<string, { ... }>, request_id, transaction_key } }

Run it

Every call needs your API token in LOOOT_TOKEN; Sign in shows how to get one. Each run also needs a new idempotency key, so a retry never pays twice. With wait: 30 the answer comes back inline when the run ends within 30 seconds. Otherwise you get the running run back: poll GET /v1/runs/<runId>.

Example input with placeholder values:

curl -X POST "https://api.looot.ai/v1/runs" \
  -H "Authorization: Bearer $LOOOT_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{"endpointId":"job:audio.transcribe","input":{"url":"https://example.com"},"wait":30}'

const response = await fetch("https://api.looot.ai/v1/runs", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.LOOOT_TOKEN}`,
    "Content-Type": "application/json",
    "Idempotency-Key": crypto.randomUUID(),
  },
  body: JSON.stringify({
    endpointId: "job:audio.transcribe",
    input: {
      url: "https://example.com",
    },
    wait: 30,
  }),
});
const run = await response.json();
console.log(run.status, run.result);

import os
import uuid

import requests

response = requests.post(
    "https://api.looot.ai/v1/runs",
    headers={
        "Authorization": f"Bearer {os.environ['LOOOT_TOKEN']}",
        "Idempotency-Key": str(uuid.uuid4()),
    },
    json={
        "endpointId": "job:audio.transcribe",
        "input": {
            "url": "https://example.com",
        },
        "wait": 30,
    },
    timeout=90,
)
run = response.json()
print(run["status"], run.get("result"))

To pin one provider, send its tool id as endpointId instead of job:audio.transcribe.