metadata
language:
- zh
license: apache-2.0
base_model: openai/whisper-tiny
tags:
- generated_from_trainer
datasets:
- formospeech/hat_asr_aligned
model-index:
- name: Whisper Tiny Hakka Webcam
results: []
Whisper Tiny Hakka Webcam
This model is a fine-tuned version of openai/whisper-tiny on the HAT ASR Aligned dataset. It achieves the following results on the evaluation set:
- Loss: 0.0420
- Cer: 1.6564
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: 0.0001
- train_batch_size: 64
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1521
- training_steps: 15210
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Cer |
|---|---|---|---|---|
| 0.134 | 0.9993 | 1521 | 0.1394 | 9.3811 |
| 0.0675 | 1.9987 | 3042 | 0.0847 | 6.3469 |
| 0.0397 | 2.9980 | 4563 | 0.0680 | 3.9092 |
| 0.0219 | 3.9974 | 6084 | 0.0581 | 3.0446 |
| 0.0118 | 4.9967 | 7605 | 0.0527 | 2.8053 |
| 0.0062 | 5.9961 | 9126 | 0.0511 | 2.5961 |
| 0.0038 | 6.9954 | 10647 | 0.0479 | 2.1037 |
| 0.002 | 7.9947 | 12168 | 0.0439 | 1.8390 |
| 0.0005 | 8.9941 | 13689 | 0.0424 | 1.7049 |
| 0.0002 | 9.9934 | 15210 | 0.0420 | 1.6564 |
Framework versions
- Transformers 4.42.3
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1