Hugging Face's logo

susmitabhatt
/
xlsr-aiish-no

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
wav2vec2
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
xlsr-aiish-no / README.md
susmitabhatt's picture
susmitabhatt
End of training
481740c verified
preview code
|
raw
history blame contribute delete
4.33 kB
metadata
library_name: transformers
license: apache-2.0
base_model: facebook/wav2vec2-large-xlsr-53
tags:
  - generated_from_trainer
metrics:
  - wer
model-index:
  - name: xlsr-aiish-no
    results: []

xlsr-aiish-no

This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0000
  • Wer: 0.3093

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: 0.0004
  • train_batch_size: 8
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 16
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 132
  • num_epochs: 100
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer
4.5121 2.1505 200 2.5138 1.0
1.2964 4.3011 400 0.1073 0.4756
0.1795 6.4516 600 0.0381 0.3814
0.096 8.6022 800 0.0067 0.3117
0.0584 10.7527 1000 0.0140 0.3227
0.0457 12.9032 1200 0.0039 0.3130
0.0412 15.0538 1400 0.0024 0.3081
0.0269 17.2043 1600 0.0142 0.3093
0.0276 19.3548 1800 0.0013 0.3068
0.0279 21.5054 2000 0.0044 0.3117
0.0243 23.6559 2200 0.0026 0.3105
0.0178 25.8065 2400 0.0006 0.3081
0.0193 27.9570 2600 0.0115 0.3215
0.0237 30.1075 2800 0.0008 0.3068
0.0146 32.2581 3000 0.0011 0.3105
0.0109 34.4086 3200 0.0002 0.3068
0.0106 36.5591 3400 0.0011 0.3081
0.0171 38.7097 3600 0.0012 0.3093
0.0099 40.8602 3800 0.0002 0.3130
0.0102 43.0108 4000 0.0014 0.3154
0.0129 45.1613 4200 0.0003 0.3105
0.0108 47.3118 4400 0.0001 0.3068
0.0085 49.4624 4600 0.0001 0.3093
0.0067 51.6129 4800 0.0001 0.3081
0.0079 53.7634 5000 0.0006 0.3068
0.0078 55.9140 5200 0.0001 0.3093
0.0091 58.0645 5400 0.0020 0.3081
0.0071 60.2151 5600 0.0017 0.3154
0.004 62.3656 5800 0.0001 0.3105
0.004 64.5161 6000 0.0001 0.3093
0.0064 66.6667 6200 0.0096 0.3166
0.0048 68.8172 6400 0.0000 0.3068
0.0037 70.9677 6600 0.0321 0.3081
0.0041 73.1183 6800 0.0000 0.3093
0.0059 75.2688 7000 0.0013 0.3093
0.0019 77.4194 7200 0.0011 0.3081
0.0022 79.5699 7400 0.0000 0.3068
0.0022 81.7204 7600 0.0000 0.3068
0.004 83.8710 7800 0.0000 0.3081
0.0025 86.0215 8000 0.0000 0.3081
0.0032 88.1720 8200 0.0000 0.3081
0.0019 90.3226 8400 0.0000 0.3093
0.001 92.4731 8600 0.0000 0.3081
0.001 94.6237 8800 0.0000 0.3093
0.0018 96.7742 9000 0.0000 0.3093
0.0019 98.9247 9200 0.0000 0.3093

Framework versions

  • Transformers 4.45.0.dev0
  • Pytorch 2.4.0
  • Datasets 2.21.0
  • Tokenizers 0.19.1