Be part of our every day and weekly newsletters for the most recent updates and unique content material on industry-leading AI protection. Study Extra
Nvidia has change into one of the worthwhile corporations on the earth lately because of the inventory market noticing how a lot demand there may be for graphics processing items (GPUs), the highly effective chips Nvidia makes which can be used to render graphics in video video games but additionally, more and more, practice AI massive language and diffusion fashions.
However Nvidia does way over simply make {hardware}, after all, and the software program to run it. Because the generative AI period wears on, the Santa Clara-based firm has additionally been steadily releasing increasingly more of its personal AI fashions — largely open supply and free for researchers and builders to take, obtain, modify and use commercially — and the most recent amongst them is Parakeet-TDT-0.6B-v2, an automated speech recognition (ASR) mannequin that may, in the phrases of Hugging Face’s Vaibhav “VB” Srivastav, “transcribe 60 minutes of audio in 1 second [mind blown emoji].”
That is the brand new technology of the Parakeet mannequin Nvidia first unveiled again in January 2024 and up to date once more in April of that 12 months, however this model two is so highly effective, it at present tops the Hugging Face Open ASR Leaderboard with a mean “Phrase Error Price” (instances the mannequin incorrectly transcribes a spoken phrase) of simply 6.05% (out of 100).
To place that in perspective, it nears proprietary transcription fashions reminiscent of OpenAI’s GPT-4o-transcribe (with a WER of two.46% in English) and ElevenLabs Scribe (3.3%).
And it’s providing all this whereas remaining freely accessible below a commercially permissive Inventive Commons CC-BY-4.0 license, making it a pretty proposition for industrial enterprises and indie builders trying to construct speech recognition and transcription providers into their paid functions.
Efficiency and benchmark standing
The mannequin boasts 600 million parameters and leverages a mixture of the FastConformer encoder and TDT decoder architectures.
It’s able to transcribing an hour of audio in only one second, supplied it’s operating on Nvidia’s GPU-accelerated {hardware}.
The efficiency benchmark is measured at an RTFx (Actual-Time Issue) of 3386.02 with a batch dimension of 128, inserting it on the prime of present ASR benchmarks maintained by Hugging Face.
Use circumstances and availability
Launched globally on Could 1, 2025, Parakeet-TDT-0.6B-v2 is geared toward builders, researchers, and {industry} groups constructing functions reminiscent of transcription providers, voice assistants, subtitle mills, and conversational AI platforms.
The mannequin helps punctuation, capitalization, and detailed word-level timestamping, providing a full transcription package deal for a variety of speech-to-text wants.
Entry and deployment
Builders can deploy the mannequin utilizing Nvidia’s NeMo toolkit. The setup course of is appropriate with Python and PyTorch, and the mannequin can be utilized immediately or fine-tuned for domain-specific duties.
The open-source license (CC-BY-4.0) additionally permits for industrial use, making it interesting to startups and enterprises alike.
Coaching information and mannequin improvement
Parakeet-TDT-0.6B-v2 was skilled on a various and large-scale corpus known as the Granary dataset. This contains round 120,000 hours of English audio, composed of 10,000 hours of high-quality human-transcribed information and 110,000 hours of pseudo-labeled speech.
Sources vary from well-known datasets like LibriSpeech and Mozilla Frequent Voice to YouTube-Commons and Librilight.
Nvidia plans to make the Granary dataset publicly accessible following its presentation at Interspeech 2025.
Analysis and robustness
The mannequin was evaluated throughout a number of English-language ASR benchmarks, together with AMI, Earnings22, GigaSpeech, and SPGISpeech, and confirmed sturdy generalization efficiency. It stays strong below diverse noise situations and performs effectively even with telephony-style audio codecs, with solely modest degradation at decrease signal-to-noise ratios.
{Hardware} compatibility and effectivity
Parakeet-TDT-0.6B-v2 is optimized for Nvidia GPU environments, supporting {hardware} such because the A100, H100, T4, and V100 boards.
Whereas high-end GPUs maximize efficiency, the mannequin can nonetheless be loaded on techniques with as little as 2GB of RAM, permitting for broader deployment eventualities.
Moral concerns and accountable use
NVIDIA notes that the mannequin was developed with out the usage of private information and adheres to its accountable AI framework.
Though no particular measures had been taken to mitigate demographic bias, the mannequin handed inside high quality requirements and contains detailed documentation on its coaching course of, dataset provenance, and privateness compliance.
The discharge drew consideration from the machine studying and open-source communities, particularly after being publicly highlighted on social media. Commentators famous the mannequin’s capability to outperform industrial ASR options whereas remaining absolutely open supply and commercially usable.
Builders taken with attempting the mannequin can entry it by way of Hugging Face or by Nvidia’s NeMo toolkit. Set up directions, demo scripts, and integration steering are available to facilitate experimentation and deployment.
