Cheapest transcription API.
prices as of · re-ranked daily · 4 qualifying models · prices are per hour of audio
Every model that turns recorded audio into text and has a price we can hold beside the others, cheapest first. 4 qualify, from $0.22 an hour. Where a lab prices several variants of one model the same, one of them stands for the rest. Prices are for one hour of recorded audio transcribed in batch on the standard model.
Today's pick is Scribe 1 at ElevenLabs, $0.22 per hour of audio, read on 17 Sept 2026.
- 1
ElevenLabsScribe 1
sold by 1 provider, $0.22 per hour of audio at ElevenLabsThe price ElevenLabs charges for the model it trained. Every row below is measured against it.
$0.22 per hour of audio
- 2
DeepgramNova 2
sold by 1 provider, $0.258 per hour of audio at DeepgramThe price Deepgram charges for the model it trained. Every row below is measured against it.
$0.258 per hour of audio
- 3Deepgram
Nova 3 Medical
sold by 1 provider, $0.312 per hour of audio at DeepgramThe price Deepgram charges for the model it trained. Every row below is measured against it.
$0.312 per hour of audio
- 4
Azure AI FoundryWhisper
sold by 3 providers, $0.36 per hour of audio at Azure AI FoundryThe same price OpenAI charges.
$0.36 per hour of audio
Frequently asked
- What is the cheapest of the transcription models right now?
- Scribe 1 at ElevenLabs, $0.22 per hour of audio, read on 17 Sept 2026.
- Which lab shows up most in this ranking?
- Deepgram trained 2 of the top 4 models here.
- How is this ranked, and how current is it?
- Every row is the cheapest price for that model that we can hold beside the others: the lab's own, a price below it with a published reason, or a price at or above it. A price that buys a different task, or a resolution the list price does not state, or a unit we will not convert without publishing the assumption, is shown on the model's page and ranked nowhere. Prices are read daily from 7 providers' public catalogs, against a market median of $0.258 per hour of audio. Last refresh: 17 Sept 2026.