metadata

library_name: transformers
license: apache-2.0
base_model: mouseyy/result_data-1
tags:
  - generated_from_trainer
datasets:
  - common_voice_17_0
metrics:
  - wer
model-index:
  - name: result_data_2-3
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: common_voice_17_0
          type: common_voice_17_0
          config: uk
          split: test
          args: uk
        metrics:
          - name: Wer
            type: wer
            value: 0.3390885932635154

result_data_2-3

This model is a fine-tuned version of mouseyy/result_data-1 on the common_voice_17_0 dataset. It achieves the following results on the evaluation set:

Loss: 0.2363
Wer: 0.3391
Cer: 0.1637

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: 4.1584533148786865e-05
train_batch_size: 16
eval_batch_size: 16
seed: 42
distributed_type: multi-GPU
num_devices: 2
total_train_batch_size: 32
total_eval_batch_size: 32
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 135
num_epochs: 5.0
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
0.1644	0.9099	1000	0.2423	0.3691	0.1712
0.1373	1.8198	2000	0.2359	0.3586	0.1699
0.1219	2.7298	3000	0.2351	0.3487	0.1665
0.1106	3.6397	4000	0.2399	0.3447	0.1667
0.0986	4.5496	5000	0.2418	0.3388	0.1639

Framework versions

Transformers 4.49.0
Pytorch 2.6.0+cu124
Datasets 3.3.2
Tokenizers 0.21.0