CoolFace
Modelpublic

gokul-g-menon/XLSR-LM-NewData

sourceHugging Faceapache-2.0updated 2y agoView on Hugging Face
0likes12downloads
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. -->

XLSR-LM-NewData

This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the None dataset. It achieves the following results on the evaluation set:

  • —Loss: 0.1395
  • —Wer: 0.2952

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: 0.00024
  • —trainbatchsize: 1
  • —evalbatchsize: 8
  • —seed: 42
  • —gradientaccumulationsteps: 4
  • —totaltrainbatch_size: 4
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: linear
  • —lrschedulerwarmup_steps: 800
  • —num_epochs: 1
  • —mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation LossWer
1.49120.116510000.54970.7011
0.53770.233020000.32920.5364
0.43430.349430000.24750.4424
0.36780.465940000.21450.4014
0.33450.582450000.18980.3774
0.30290.698960000.17180.3441
0.26850.815370000.15170.3135
0.23850.931880000.13950.2952

Framework versions

  • —Transformers 4.42.4
  • —Pytorch 2.3.1+cu121
  • —Datasets 2.20.0
  • —Tokenizers 0.19.1