Edit model card

w2v2-ks-jpqd-quant-FE-finetuned-student

This model is a fine-tuned version of anton-l/wav2vec2-base-ft-keyword-spotting on the superb dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0869
  • Accuracy: 0.9794

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: 7e-05
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 128
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.5
  • num_epochs: 12.0
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy
0.3477 1.0 399 0.1516 0.9637
5.5957 2.0 798 5.4798 0.9545
8.7806 3.0 1197 8.6491 0.9634
10.4524 4.0 1596 10.2701 0.9554
10.8964 5.0 1995 10.7809 0.9647
10.9322 6.0 2394 10.7806 0.9619
0.2389 7.0 2793 0.1148 0.9738
0.2522 8.0 3192 0.1013 0.9747
0.2213 9.0 3591 0.0983 0.9754
0.2053 10.0 3990 0.0934 0.9768
0.1543 11.0 4389 0.0875 0.9779
0.1836 12.0 4788 0.0869 0.9794

Framework versions

  • Transformers 4.26.0
  • Pytorch 1.13.1+cu116
  • Datasets 2.8.0
  • Tokenizers 0.13.2
Downloads last month
5

Dataset used to train yujiepan/internal.wav2vec2-base-superb-ks-int8-structured64-quantize-feature-extractor