metadata
library_name: transformers
language:
- hi
license: apache-2.0
base_model: openai/whisper-small
tags:
- hf-asr-leaderboard
- generated_from_trainer
datasets:
- aihub_adult_baseline
model-index:
- name: whisper-small-E10_Yfreq_speed
results: []
whisper-small-E10_Yfreq_speed
This model is a fine-tuned version of openai/whisper-small on the aihub old adult freq speed pause changed dataset. It achieves the following results on the evaluation set:
- Loss: 0.2283
- Cer: 6.1736
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: 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: 50
- num_epochs: 2
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Cer |
|---|---|---|---|---|
| 0.2612 | 0.1289 | 100 | 0.2864 | 7.4836 |
| 0.1602 | 0.2579 | 200 | 0.2540 | 6.8492 |
| 0.126 | 0.3868 | 300 | 0.2559 | 6.9960 |
| 0.1277 | 0.5158 | 400 | 0.2460 | 7.0019 |
| 0.1111 | 0.6447 | 500 | 0.2426 | 6.7434 |
| 0.1045 | 0.7737 | 600 | 0.2386 | 6.4262 |
| 0.1021 | 0.9026 | 700 | 0.2377 | 6.3029 |
| 0.0431 | 1.0309 | 800 | 0.2263 | 6.1090 |
| 0.0314 | 1.1599 | 900 | 0.2312 | 6.1443 |
| 0.0389 | 1.2888 | 1000 | 0.2326 | 6.3029 |
| 0.0366 | 1.4178 | 1100 | 0.2314 | 6.0503 |
| 0.0403 | 1.5467 | 1200 | 0.2315 | 6.1031 |
| 0.0374 | 1.6757 | 1300 | 0.2286 | 5.9915 |
| 0.0339 | 1.8046 | 1400 | 0.2278 | 6.0503 |
| 0.0385 | 1.9336 | 1500 | 0.2283 | 6.1736 |
Framework versions
- Transformers 4.47.0.dev0
- Pytorch 2.5.0+cu121
- Datasets 3.1.0
- Tokenizers 0.20.3