Voice and music · by NVIDIA

Nemotron Transcribe

Transcribe or ask questions about hour-long recordings in 40 languages.

speech → text, data · speech, text → text
Way in
Official APISign up yourself, api key
To start
$open weights (OpenMDW-1.1)
Runs
In the cloudAlso through deepinfra, Hugging Face, nvidia-nim
Checked
Nov 4, 2026Prices as of that day

What it can do 2

Transcribe up to 1 h of audio in 40 languages, with optional word timestamps and speaker attributionself-hosting needs 1x H100 80 GB (FP8/NVFP4 variants are smaller)speech → text, data$0.002/min on DeepInfra; free self-hosted
Translate speech or answer questions about an audio recordingspeech, text → textsame

Does the same job

Prices and limits as of Nov 4, 2026, from NVIDIA’s own pages. They change; check before you rely on them.Not affiliated with NVIDIA. Not advice.Something wrong?
Made by Jasper Burns

A short note when I ship something new, about once a month. One list for all my projects.