NVIDIA Parakeet: fast open speech recognition

by NVIDIA

Don’t want to install NVIDIA Parakeet? Drop the file here.

The first 20 minutes of any recording are transcribed free: a recording up to 20 minutes transcribes in full, a longer one is previewed, and a pack from $6 (500 minutes) transcribes the rest and opens every export, and credits never expire.

parakeet-tdt-0.6b-v3: 600 million parameters, 25 European languages detected automatically, punctuation, word timestamps, and audio up to 3 hours with local attention.

Linux  ·  free

Get NVIDIA Parakeet Official site & pricing

TL;DR

Parakeet TDT 0.6B v3: 600M parameters, 25 European languages auto-detected, punctuation and word timestamps, up to 3 hours of audio. CC-BY-4.0, commercial use allowed.

It powers local apps like Handy, OpenWhispr and Whisper Notes. The toolkit behind it is NeMo.

Category
Open source
License
CC-BY-4.0
Pricing
free (CC-BY-4.0)
Platforms
Linux

What it is

Parakeet is NVIDIA's family of fast ASR models in the NeMo toolkit. parakeet-tdt-0.6b-v3 (August 2025) has 600 million parameters and extends the English-only v2 to 25 European languages, including Finnish and Latvian, detecting the language with no prompt. It outputs punctuation and capitalisation with word- and segment-level timestamps, and handles audio up to 24 minutes with full attention on an A100 80GB or up to 3 hours with local attention. It is CC-BY-4.0 and the card states it is ready for commercial use. Desktop apps such as Handy, OpenWhispr, VoiceInk, Voicetypr, Whisper Notes and Scriberr ship it as a local engine.

Best for: Developers who want a fast, small, commercially usable model for English and European languages, on a GPU or inside a local app.
Watch out for: 25 European languages only (v2 is English-only); no Asian, Middle Eastern or African languages. Full attention handles about 24 minutes on an A100 80GB; longer audio needs local attention.

Transcripts without a GPU or NeMo

  1. Drop the audio or video file into the box at the top of this page, or paste a link (YouTube, a podcast or RSS feed, Google Drive, Twitch or Kick). Nothing to install and no GPU needed.
  2. The language is detected automatically and the transcript stays in the language that was spoken (WhipScribe does not translate). That includes languages outside Parakeet's 25.
  3. Read the summary free. A pack from $6 (500 minutes) opens the full transcript and the downloads: TXT, DOCX, SRT, VTT and JSON. Packs are one-time purchases and credits never expire.
  4. Files can be up to 10 hours or 5 GB per file.

Related: Bulk transcription · SRT generator · Transcribe large files · WhisperX · Pricing

Install / use

pip install -U "nemo_toolkit[asr]"  # model: nvidia/parakeet-tdt-0.6b-v3

Key specs

Parameters
600M
Languages
25 European (auto-detected)
Max audio
24 min full / 3 h local attention
Licence
CC-BY-4.0

What people actually do with NVIDIA Parakeet-style transcription

The tool is the means. These are the jobs — each one priced at published rates, each one wired up on its own page.

Wire it into Amazon S3n8nMake.comGoogle DriveZapierSlack

Features

Speaker diarizationNot stated
Word-level timestampsYes
Streaming / real-timeNot stated
Languages supported25
HIPAA eligibleNo

Links

NVIDIA Parakeet vs Whipscribe

FeatureNVIDIA ParakeetWhipscribe
CategoryOpen sourceTranscription APIs
Pricingfree (CC-BY-4.0)$6–$24 one-time packs (credits never expire) · first 20 minutes free
Speaker identificationNoNamed in the summary, key quotes and chapters
Word timestampsYesYes
StreamingNoNo
Languages2599
PlatformsLinuxWeb, API, MCP

Alternatives to NVIDIA Parakeet

Frequently asked about NVIDIA Parakeet

Which languages does Parakeet v3 support?

25 European languages: Bulgarian, Croatian, Czech, Danish, Dutch, English, Estonian, Finnish, French, German, Greek, Hungarian, Italian, Latvian, Lithuanian, Maltese, Polish, Portuguese, Romanian, Russian, Slovak, Slovenian, Spanish, Swedish and Ukrainian. parakeet-tdt-0.6b-v2 is English-only.

How long a file can Parakeet transcribe?

Per the model card, up to 24 minutes with full attention on an A100 80GB, or up to 3 hours with local attention.

Can I use Parakeet commercially?

Yes. The model is released under CC-BY-4.0 and the card states it is ready for commercial and non-commercial use; attribution is required.

Does Parakeet give word timestamps?

Yes, at character, word and segment level, with punctuation and capitalisation in the text.

Which apps use Parakeet?

Among the apps on this site: Handy, OpenWhispr, VoiceInk, Voicetypr, Whisper Notes, Scriberr and Whispering ship Parakeet as a local engine.

What is NVIDIA Parakeet?

Parakeet TDT 0.6B v3: 600M parameters, 25 European languages auto-detected, punctuation and word timestamps, up to 3 hours of audio. CC-BY-4.0, commercial use allowed.

Is NVIDIA Parakeet free?

Yes. NVIDIA Parakeet is free and open source (free (CC-BY-4.0)).

Does NVIDIA Parakeet work on Mac, Windows and Linux?

NVIDIA Parakeet runs on Linux. It is not listed for other platforms; check the vendor for anything beyond Linux.

How do I install NVIDIA Parakeet?

With pip: `pip install -U "nemo_toolkit[asr]" # model: nvidia/parakeet-tdt-0.6b-v3`. You need Python and, for most audio tools, ffmpeg on your PATH first.

Is NVIDIA Parakeet HIPAA compliant?

NVIDIA Parakeet does not claim HIPAA compliance. For PHI, run it on infrastructure you control or choose a service that will sign a BAA.

What are the limitations of NVIDIA Parakeet?

25 European languages only (v2 is English-only); no Asian, Middle Eastern or African languages. Full attention handles about 24 minutes on an A100 80GB; longer audio needs local attention.

Who is NVIDIA Parakeet best for?

Developers who want a fast, small, commercially usable model for English and European languages, on a GPU or inside a local app.

What should I know before choosing NVIDIA Parakeet?

It powers local apps like Handy, OpenWhispr and Whisper Notes. The toolkit behind it is NeMo.

Does NVIDIA Parakeet run offline?

NVIDIA Parakeet runs on your own machine — it is open source, so nothing leaves the computer unless you configure it to.

Can I use NVIDIA Parakeet in the browser?

No — NVIDIA Parakeet is a Linux application, not a web app. You install it rather than sign in to it.

Whipscribe is a hosted transcription service. If you want transcripts without running infrastructure, paste a URL or drop a file in the form below — you'll have a transcript in seconds.

Transcribe your recordings

Explore

Long recording to get through?

One file up to 10 hours, as an upload or a link: see how to transcribe an MP3 of any length, how to transcribe a podcast and the SRT file generator.

Rather have it done for you?

Free tools cost setup time. When you want a finished transcript instead: see how the best transcription services compare, cheap transcription services and the real cost per hour and transcription services from $6.