CoolFace
Modelpublic

Sandiago21/whisper-tiny-PolyAI-minds14

sourceHugging Faceapache-2.0updated 3y agoView on Hugging Face
1likes20downloads
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-tiny-finetuned-minds14

This model is a fine-tuned version of openai/whisper-tiny on the MINDS14 dataset. It achieves the following results on the evaluation set:

  • —Loss: 0.6435
  • —Wer Ortho: 0.3797
  • —Wer: 0.3499

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: 1e-05
  • —trainbatchsize: 16
  • —evalbatchsize: 16
  • —seed: 42
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: linear
  • —lrschedulerwarmup_steps: 50
  • —num_epochs: 10

Training results

Training LossEpochStepValidation LossWer OrthoWer
4.9951.0292.98790.54250.4127
2.16342.0580.80840.43820.3936
0.66593.0870.62680.41440.3678
0.38654.01160.59870.38800.3561
0.24285.01450.60050.39900.3659
0.17346.01740.61620.39060.3573
0.09657.02030.62210.38930.3561
0.06828.02320.63200.38030.3493
0.04739.02610.64110.37970.3493
0.047610.02900.64350.37970.3499

Framework versions

  • —Transformers 4.30.0.dev0
  • —Pytorch 2.0.1+cu117
  • —Datasets 2.13.1
  • —Tokenizers 0.13.3