NVIDIA
Explore
Models
Blueprints
GPUs
Docs
⌘KCtrl+K
Terms of Use
Privacy Policy
Your Privacy Choices
Contact

Copyright © 2026 NVIDIA Corporation

nvidia

parakeet-1.1b-rnnt-multilingual-asr

Downloadable

High accuracy and optimized performance for transcription in 25 languages

Automatic Speech RecognitionNVIDIA NIMNVIDIA Riva
Get API Key
API ReferenceAPI Reference
Accelerated by DGX Cloud

Speech Recognition: Parakeet

Description

NVIDIA Parakeet‑RNNT‑1.1B is a multilingual ASR model with 1.1B parameters, trained on more than 90K hours of speech for high‑quality transcription across global use cases. It uses a universal tokenizer built by first training language‑specific tokenizers and then merging them into a single shared vocabulary, enabling efficient cross‑lingual learning and deployment.

The model supports transcription in 25 languages and variants—including English (en‑US, en‑GB), Spanish (es‑US, es‑ES), German (de‑DE), French (fr‑FR, fr‑CA), Italian (it‑IT), Arabic (ar‑AR), Japanese (ja‑JP), Korean (ko‑KR), Portuguese (pt‑BR, pt‑PT), Russian (ru‑RU), Hindi (hi‑IN), Dutch (nl‑NL), Danish (da‑DK), Norwegian Nynorsk (nn‑NO), Norwegian Bokmål (nb‑NO), Czech (cs‑CZ), Polish (pl‑PL), Swedish (sv‑SE), Thai (th‑TH), Turkish (tr‑TR), and Hebrew (he‑IL).

This model produces case‑sensitive transcripts (upper and lower case) with punctuation, spaces, and apostrophes, making it suitable for production‑grade voice applications and downstream language understanding.
This model is ready for commercial use.

License/Terms of Use

GOVERNING TERMS: The use of this model is governed by the NVIDIA Community Model License (found at https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-community-models-license/).

References

[1] Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition
[2] Fast-Conformer-RNNT Model
[3] Conformer: Convolution-augmented Transformer for Speech Recognition

Model Architecture

Architecture Type: Parakeet-RNNT (also known as FastConformer-RNNT) [1], [2] which is an optimized version of Conformer model [3] with 8x depthwise-separable convolutional downsampling with Hybrid loss
Network Architecture: Parakeet-RNNT-XXL-1.1B

Input

Input Type(s): Audio
Input Format(s): wav
Input Parameters: 1-Dimension
Other Properties Related to Input: Maximum Length in seconds specific to GPU Memory, No Pre-Processing Needed, Mono channel is required

Output

Output Type(s): Text
Output Format: String
Output Parameters: 1-Dimension
Other Properties Related to Output: No Maximum Character Length, Does not handle special characters

Software Integration

Supported Hardware Microarchitecture Compatibility:

  • NVIDIA Blackwell
  • NVIDIA Ampere
  • NVIDIA Hopper
  • NVIDIA Jetson
  • NVIDIA Turing

[Preferred/Supported] Operating System(s):

  • Linux

Model Version(s):

Parakeet-RNNT-XXL-1.1b_universal_spe8.5k_1.0

Inference

Engine: Triton
Test Hardware:

  • NVIDIA A10
  • NVIDIA A100
  • NVIDIA A30
  • NVIDIA H100
  • NVIDIA L4
  • NVIDIA L40
  • NVIDIA Turing T4

Ethical Considerations (For NVIDIA Models Only):

NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications. When downloaded or used in accordance with our terms of service, developers should work with their internal model team to ensure this model meets requirements for the relevant industry and use case and addresses unforeseen product misuse. For more detailed information on ethical considerations for this model, please see the Model Card++ Explainability, Bias, Safety & Security, and Privacy Subcards. Please report security vulnerabilities or NVIDIA AI Concerns here.