Skip to main content
NVIDIA Deep Learning Examples
Jasper checkpoint (PyTorch, AMP, LibriSpeech)
Model
NVIDIA Deep Learning Examples
Jasper checkpoint (PyTorch, AMP, LibriSpeech)

Jasper PyTorch checkpoint trained on LibriSpeech (test-other 9.66% WER)

2 Versions
02/10/2021 12:37 AM UTC1.24 GBAccuracy: 0440 EpochsBatch Size: 512GPU: A100
architecture
KeyValue
typejasper10x5dr_speedp-online_speca: SpecAugment (2x20, 3x75); Speed perturbation U(0.85, 1.15)
performance
KeyValue
dev_clean_wer3.20
dev_other_wer9.74
test_clean_wer3.38
test_other_wer9.66
training
KeyValue
batch_size512
datasetLibriSpeech
epochs440
training_precisionAMP
19.08.0Selected
08/21/2020 1:36 AM UTC2.48 GBAccuracy: 0400 EpochsBatch Size: 0GPU: V100
architecture
KeyValue
dense_residual_connnectionsTrue
layers50 conv layers - 10 blocks - each 5 subblocks
model_configsjasper10x5dr_sp_offline_specaugment.toml
performance
KeyValue
WER (on Librispeech dev-clean)<3.64%
training
KeyValue
bs512
cudnnbenchmark
epochs400
gpuV100
gpus8
grad_acc_steps1
lr0.015
lr_decayquadratic
optimnovograd
spec_augTrue
speed_perturb1.0;0.9;1.1
wd1e-3
1-2 of 2 items

Modal Content

NVIDIA uses cookies to improve your experience on our web site. We and our third-party partners also use cookies and other tools to collect and record information you provide as well as information about your interactions with our websites for performance improvement, analytics, and to assist in marketing efforts. By clicking "Accept All", you consent to our use of cookies and other tools as described in our Cookie Policy. You can manage your cookie settings by clicking on "Manage Settings." By continuing to use this site or by clicking one of the buttons below, you agree to our Terms of Service (which contains important waivers). Please see our Privacy Policy for more information on our privacy practices.