CatalogModels
Models
The NGC catalog offers 100s of pre-trained models for computer vision, speech, recommendation, and more. Bring AI faster to market by using these models as-is or quickly build proprietary models with a fraction of your custom data.
Sort: Last Modified
Spanish Conformer ASR model trained on ASR set 2.0.
ResNet50 ImageNet pretrained weights.
PeopleNet
Model
3 class object detection network to detect people in an image.
Pretrained weights to facilitate transfer learning using TAO Toolkit.
Mandarin (zh-CN) Conformer ASR model trained on ASR set 5.0
German Conformer ASR model trained on RIVA ASR set
French Conformer ASR model trained on RIVA ASR set
English (en-GB) Conformer ASR model trained on ASR set 1.0
StyleGAN3 pretrained models for FFHQ, AFHQv2 and MetFaces datasets.
For use with the official StyleGAN3 implementation: https://github.com/NVlabs/stylegan3
TTS En HiFiTTS VITS
Model
End-to-end parallel speech synthesis model
Base German grammar
Base Spanish grammar
Base English grammar
Base French grammar
Riva multisepaker with IPA for G2P
FastPitch is a mel-spectrogram generator, designed to be used as the first part of a neural text-to-speech system in conjunction with a neural vocoder
Mel-Spectrogram prediction conditioned on input text with LJSpeech voice.
Riva multisepaker with IPA for G2P
GAN-based waveform generator from mel-spectrograms.
HifiGAN is a neural vocoder model for text-to-speech applications. It is intended as the second part of a two-stage speech synthesis pipeline, with a mel-spectrogram generator such as FastPitch as the first stage.
RIVA ASR Russian LM
Model
Base Russian n-gram LM
Riva multisepaker with IPA for G2P
Riva multisepaker with IPA for G2P
RIVA Punctuation
Model