CoolFace
Modelpublic

ales/whisper-base-belarusian

sourceHugging Faceapache-2.0updated 4y agoView on Hugging Face
2likes13downloads
Model Card

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

Whisper Base Belarusian

This model is a fine-tuned version of openai/whisper-base on the mozilla-foundation/commonvoice11_0 be dataset. It achieves the following results on the evaluation set:

  • —Loss: 0.1080
  • —Wer: 12.2069

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: 0.0001
  • —trainbatchsize: 64
  • —evalbatchsize: 32
  • —seed: 42
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: linear
  • —lrschedulerwarmup_steps: 500
  • —training_steps: 6000
  • —mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation LossWer
0.24450.1710000.305932.4163
0.18230.3320000.200422.1259
0.14120.530000.175220.0700
0.10930.6740000.141316.0533
0.11370.8350000.115513.3108
0.05851.160000.108012.2069

Framework versions

  • —Transformers 4.26.0.dev0
  • —Pytorch 1.13.0+cu117
  • —Datasets 2.7.1.dev0
  • —Tokenizers 0.13.2