metadata

language:
  - ko
license: apache-2.0
tags:
  - hf-asr-leaderboard
  - generated_from_trainer
base_model: openai/whisper-small
datasets:
  - Marcusxx/CngFSt3T
model-index:
  - name: CngFSt3T_small_model
    results: []

CngFSt3T_small_model

This model is a fine-tuned version of openai/whisper-small on the Marcusxx/CngFSt3T dataset. It achieves the following results on the evaluation set:

Loss: 0.0723
Cer: 135.0080

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: 1e-05
train_batch_size: 16
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 250
training_steps: 5000
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Cer
0.0452	2.2321	1000	0.1419	74.5039
0.0061	4.4643	2000	0.0770	130.8381
0.0006	6.6964	3000	0.0710	118.6804
0.0003	8.9286	4000	0.0722	121.9459
0.0003	11.1607	5000	0.0723	135.0080

Framework versions

Transformers 4.41.2
Pytorch 2.2.2+cu121
Datasets 2.19.2
Tokenizers 0.19.1