Vendor page
nvidia/parakeet-tdt-0.6b-v3 model card (Hugging Face)
NVIDIA, Hugging Face, 2025
Read the originalRead on October 16, 2026
- What it is
- The model card for NVIDIA's parakeet-tdt-0.6b-v3 speech recognition model. It gives the size, the 25 European languages, the license and the hardware notes.
- How much weight it carries
- The model publisher's own card. It states the model's design and license; the hardware line is NVIDIA's view and differs from some app descriptions of the same model.
Facts taken from this source
- NVIDIA's model card for Parakeet TDT 0.6B v3 says its transcripts may be less than 100 percent accurate, and that accuracy depends on the language and the characteristics of the input audio.
“Accuracy varies based on language and characteristics of input audio” (Technical Limitations and Mitigation section of the model card)
- NVIDIA's model card for Parakeet TDT 0.6B v3 says that to run Python inference with the model, you install NVIDIA NeMo after installing a recent PyTorch version.
“To train, fine-tune, or run Python inference with this model, install NVIDIA NeMo after installing a recent PyTorch version.” (NVIDIA NeMo section of the model card)
- NVIDIA says Parakeet TDT 0.6B v3 handles audio up to 24 minutes long with full attention, or up to 3 hours with local attention.
“supporting audio up to 24 minutes long with full attention (on A100 80GB) or up to 3 hours with local attention” (Key features, long audio transcription)
- NVIDIA says the parakeet-tdt-0.6b-v3 model extends language support from English to 25 European languages.
“expanding language support from English to 25 European languages” (Description)
- NVIDIA's parakeet-tdt-0.6b-v3 is a multilingual speech recognition model with 600 million parameters.
“is a 600-million-parameter multilingual automatic speech recognition (ASR) model” (Description)
- NVIDIA states that use of the parakeet-tdt-0.6b-v3 model is governed by the CC-BY-4.0 license.
“Use of this model is governed by the CC-BY-4.0 license” (License/Terms of Use)
- NVIDIA says its models are designed and optimized for NVIDIA GPU-accelerated systems.
“designed and/or optimized to run on NVIDIA GPU-accelerated systems” (Hardware section)
Pages that rest on this source
- Speech Recognition in Python: Libraries ComparedGuide
- Open-Source Speech Recognition: The Best ModelsGuide
- Dictation Software: How to Choose the Right AppTopic
- Dictation on Windows: Voice Typing, Voice Access, AppsTopic
- Speech Recognition: What It Is and How Well It WorksTopic
- Open-Source Dictation Software: Free Apps on GitHubGuide
- Speech to Text in French, German, Russian and MoreGuide
- What are the best alternatives to Whisper?Question
- What is NVIDIA Parakeet, and is it better than Whisper?Question
- Offline Speech to Text on Windows: What WorksGuide
- Which open-source dictation apps run on Windows?Question