patrickvonplaten / wavlm-libri-clean-100h-base

huggingface.co
Total runs: 23
24-hour runs: 0
7-day runs: -7
30-day runs: -937
Model's Last Updated: December 20 2021
automatic-speech-recognition

Introduction of wavlm-libri-clean-100h-base

Model Details of wavlm-libri-clean-100h-base

wavlm-libri-clean-100h-base

This model is a fine-tuned version of microsoft/wavlm-base on the LIBRISPEECH_ASR - CLEAN dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0829
  • Wer: 0.0675
Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure
Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 0.0003
  • train_batch_size: 4
  • eval_batch_size: 4
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 8
  • total_train_batch_size: 32
  • total_eval_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 500
  • num_epochs: 3.0
  • mixed_precision_training: Native AMP
Training results
Training Loss Epoch Step Validation Loss Wer
2.8805 0.34 300 2.8686 1.0
0.2459 0.67 600 0.1858 0.1554
0.1114 1.01 900 0.1379 0.1191
0.0867 1.35 1200 0.1130 0.0961
0.0698 1.68 1500 0.1032 0.0877
0.0663 2.02 1800 0.0959 0.0785
0.0451 2.35 2100 0.0887 0.0748
0.0392 2.69 2400 0.0859 0.0698
Framework versions
  • Transformers 4.15.0.dev0
  • Pytorch 1.9.0+cu111
  • Datasets 1.16.2.dev0
  • Tokenizers 0.10.3

Runs of patrickvonplaten wavlm-libri-clean-100h-base on huggingface.co

23
Total runs
0
24-hour runs
-3
3-day runs
-7
7-day runs
-937
30-day runs

More Information About wavlm-libri-clean-100h-base huggingface.co Model

wavlm-libri-clean-100h-base huggingface.co

wavlm-libri-clean-100h-base huggingface.co is an AI model on huggingface.co that provides wavlm-libri-clean-100h-base's model effect (), which can be used instantly with this patrickvonplaten wavlm-libri-clean-100h-base model. huggingface.co supports a free trial of the wavlm-libri-clean-100h-base model, and also provides paid use of the wavlm-libri-clean-100h-base. Support call wavlm-libri-clean-100h-base model through api, including Node.js, Python, http.

wavlm-libri-clean-100h-base huggingface.co Url

https://huggingface.co/patrickvonplaten/wavlm-libri-clean-100h-base

patrickvonplaten wavlm-libri-clean-100h-base online free

wavlm-libri-clean-100h-base huggingface.co is an online trial and call api platform, which integrates wavlm-libri-clean-100h-base's modeling effects, including api services, and provides a free online trial of wavlm-libri-clean-100h-base, you can try wavlm-libri-clean-100h-base online for free by clicking the link below.

patrickvonplaten wavlm-libri-clean-100h-base online free url in huggingface.co:

https://huggingface.co/patrickvonplaten/wavlm-libri-clean-100h-base

wavlm-libri-clean-100h-base install

wavlm-libri-clean-100h-base is an open source model from GitHub that offers a free installation service, and any user can find wavlm-libri-clean-100h-base on GitHub to install. At the same time, huggingface.co provides the effect of wavlm-libri-clean-100h-base install, users can directly use wavlm-libri-clean-100h-base installed effect in huggingface.co for debugging and trial. It also supports api for free installation.

wavlm-libri-clean-100h-base install url in huggingface.co:

https://huggingface.co/patrickvonplaten/wavlm-libri-clean-100h-base

Url of wavlm-libri-clean-100h-base

wavlm-libri-clean-100h-base huggingface.co Url

Provider of wavlm-libri-clean-100h-base huggingface.co

patrickvonplaten
ORGANIZATIONS

Other API from patrickvonplaten