rossevine's picture
update model card README.md
4fe1704
|
raw
history blame
1.94 kB
metadata
license: apache-2.0
base_model: facebook/wav2vec2-xls-r-300m
tags:
  - generated_from_trainer
metrics:
  - wer
model-index:
  - name: Model_S_P_Wav2Vec2_Versi3
    results: []

Model_S_P_Wav2Vec2_Versi3

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: 2.4319
  • Wer: 0.6128
  • Cer: 0.2545

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: 16
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 2
  • total_train_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: 30

Training results

Training Loss Epoch Step Validation Loss Wer Cer
3.9619 4.17 400 2.3149 0.9821 0.4309
0.9083 8.33 800 1.8261 0.7469 0.3068
0.4814 12.5 1200 1.8669 0.7004 0.2909
0.3331 16.67 1600 2.2705 0.6705 0.2814
0.2352 20.83 2000 2.3172 0.6234 0.2607
0.1776 25.0 2400 2.2959 0.6174 0.2553
0.1122 29.17 2800 2.4319 0.6128 0.2545

Framework versions

  • Transformers 4.31.0
  • Pytorch 2.0.1+cu117
  • Datasets 1.18.3
  • Tokenizers 0.13.3