CoolFace
Modelpublic

ihanif/whisper-small-pashto-dropout

sourceHugging Faceapache-2.0updated 4y agoView on Hugging Face
0likes20downloads
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 Small Pashto

This model is a fine-tuned version of openai/whisper-small on the google/fleurs ps_af dataset. It achieves the following results on the evaluation set:

  • —Loss: 1.2273
  • —Wer: 56.6510

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: 3e-07
  • —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: 800
  • —mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation LossWer
2.11833.71001.317076.9522
0.85657.412000.936761.9930
0.224611.113000.964258.8302
0.05414.814001.087657.9903
0.015918.525001.179857.8768
0.004522.226001.230956.6510
0.0026100.07001.258156.8478
0.0023114.298001.271056.7570

Framework versions

  • —Transformers 4.26.0.dev0
  • —Pytorch 1.13.1+cu117
  • —Datasets 2.7.1.dev0
  • —Tokenizers 0.13.2