Pass IndexThe State of AISign in

Speech-to-text batch (Nova-3 monolingual)

Deepgram Nova-3 is a speech-to-text model that "represents a significant leap forward in speech AI technology", offering real-time multilingual conversation transcription, domain-specific terminology comprehension and optional personal information redaction.

audio → text · made by Deepgram

$0.0043per minute
Deepgram

Sold by 5 ways

SellerLaneRate
Deepgram apibatch$0.0043per minutedeepgram.com · read 2026-08-24
Deepgram apistreaming$0.0048per minutedeepgram.com · read 2026-08-24
Cloudflare 0 ways$0.0052$0.0092
Cloudflare aggregatornot on the seller's list todaystandard$0.0052per minutedevelopers.cloudflare.com · read 2026-08-25
Cloudflare aggregatornot on the seller's list todaywebsocket$0.0092per minutedevelopers.cloudflare.com · read 2026-08-25
Cloudflare aggregatornot on the seller's list todaywebsocket sandbox$0.0092per minutedevelopers.cloudflare.com · read 2026-08-25

About

Speech-to-text batch (Nova-3 monolingual) — a text model from Deepgram.

It takes audio and returns text. The catalogue files it under transcribe.

Every current figure

Maker
Deepgram
Register
model
Takes
audio
Returns
text

Known as 2 names

@cf/deepgram/nova-3nova-3