Make audio useful.
Transcribe interviews, podcasts, meetings, and archives for your own search and content workflows.
Turn audio into usable text. Pay by audio duration, with word-level timestamps included and no compute to reserve.
Transcribe interviews, podcasts, meetings, and archives for your own search and content workflows.
Word-level timestamps connect each part of a transcript to its place in the recording.
Submit audio through the API, then retrieve the result or receive a webhook when it’s ready.
For English speech, interviews, and recordings.
$0.05per audio hourParakeet · native jobs APIFor multilingual audio and OpenAI-compatible integrations.
$0.10per audio hourWhisper · native or OpenAI-compatible APIAvailable through reviewed pilot access. Rates in USD. Billing details →
Transcription is priced by the duration of the audio you submit. An audio hour is an hour of recording, regardless of how long it takes to process. Published rates are in USD.
No reserved infrastructure is required. Send jobs when you have work to do. Pilot metering, invoicing, and any allowances are confirmed before paid use.
Acorn-caused failures, such as a worker crash or dispatch timeout, are not billed. The API’s billed: false flag is authoritative. Malformed-input failures can incur charges for compute consumed; see the API documentation for error handling.
Explore other workloads you can run on Acorn.
Explore all products →