metadata
language:
- hi
license: apache-2.0
base_model: openai/whisper-small
tags:
- hf-asr-leaderboard
- generated_from_trainer
datasets:
- aihub_elder
model-index:
- name: whisper-small-ko-E2.1
results: []
whisper-small-ko-E2.1
This model is a fine-tuned version of openai/whisper-small on the aihub elder over 70 dataset. It achieves the following results on the evaluation set:
- Loss: 0.1594
- Cer: 4.5641
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: 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: 50
- num_epochs: 2
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Cer |
---|---|---|---|---|
0.4115 | 0.13 | 100 | 0.2656 | 6.7669 |
0.2838 | 0.26 | 200 | 0.2104 | 5.8623 |
0.2842 | 0.39 | 300 | 0.1955 | 5.3454 |
0.2437 | 0.52 | 400 | 0.1958 | 5.4922 |
0.2284 | 0.64 | 500 | 0.1828 | 5.1046 |
0.21 | 0.77 | 600 | 0.1736 | 4.7815 |
0.2066 | 0.9 | 700 | 0.1696 | 4.8637 |
0.1305 | 1.03 | 800 | 0.1670 | 4.7756 |
0.1109 | 1.16 | 900 | 0.1673 | 4.6699 |
0.1207 | 1.29 | 1000 | 0.1664 | 4.9518 |
0.1203 | 1.42 | 1100 | 0.1650 | 4.7051 |
0.1083 | 1.55 | 1200 | 0.1612 | 4.7874 |
0.1249 | 1.68 | 1300 | 0.1612 | 4.7286 |
0.1149 | 1.81 | 1400 | 0.1601 | 4.6288 |
0.1079 | 1.93 | 1500 | 0.1594 | 4.5641 |
Framework versions
- Transformers 4.37.0.dev0
- Pytorch 2.1.0+cu121
- Datasets 2.16.0
- Tokenizers 0.15.0