Deepgram / Transcription

Nova-3

Nova-3 (deepgram/nova-3) is a speech-to-text model from Deepgram on the OceanToken API at $0.0043 per minute of audio.

Deepgram Nova-3 general-purpose speech-to-text model with monolingual and multilingual transcription support.

Create an account or sign in to OceanToken.

Output

transcription

Input

audio

Price

$0.0043 / minute of audio

  • Speech-to-text
  • Meeting and call transcripts

Model ID deepgram/nova-3

Base URL https://api.oceantoken.ai/v1

One account. Your next idea.

Audio
$0.0043 / minute of audio

Official price as of October 18, 2026. The console always shows the current price.

Open OceanToken ↗Specs, request sample and alternatives in the docs ↗

Deepgram Nova-3 general-purpose speech-to-text model with monolingual and multilingual transcription support.

Nova-3 costs $0.0043 per minute of audio. This is the official price as of October 18, 2026, charged per use to your prepaid balance. The console always shows the current price.

Nova-3 takes audio as input and produces a transcript.

Send requests to https://api.oceantoken.ai/v1/audio/transcriptions with the model ID deepgram/nova-3 and your OceanToken API key. The API is OpenAI-compatible, so an OpenAI SDK needs only the base URL https://api.oceantoken.ai/v1 and your key. The same OceanToken API key works for every model your account can use. Full reference: docs.oceantoken.ai.

OceanToken is prepaid. Open Credits in the console, choose an amount and pay by card on Stripe's hosted checkout page (Link works too; Alipay and WeChat Pay are not available). Each top-up is between $10 and $100,000, and your balance updates a few seconds after payment. Have a trial code? Redeem it on the same page.

Purchased credits are valid for 12 months from the date of purchase. Credits from trial codes, promotions or gifts may have a shorter validity. All purchases are final and non-refundable, except where the law requires a refund. Section 6 of the Terms of Service has the details.

Not if the request never ran: a request refused for an insufficient balance or by a safety check, or one whose provider fails before returning any output, is not charged. A video job that fails releases the amount it held. If a request breaks off after the model has started producing output, for example a stream cut off part-way, you pay for the tokens that were processed and generated for it.

There are two ways to buy. Pay-as-you-go is self-serve: sign up, add prepaid credits and use every model in the catalog at its official price. Enterprise customers get agreement pricing based on expected usage: email support@oceantoken.ai.