metadata
language:
- multilingual
license: apache-2.0
tags:
- hf-asr-leaderboard
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_16_0
metrics:
- wer
base_model: openai/whisper-medium
model-index:
- name: Whisper medium nan-tw
results:
- task:
type: automatic-speech-recognition
name: Automatic Speech Recognition
dataset:
name: Common Voice 16.0
type: mozilla-foundation/common_voice_16_0
config: nan-tw
split: test
args: 'config: nan-tw, split: test'
metrics:
- type: wer
value: 100.17785682525566
name: Wer
Whisper medium nan-tw
This model is a fine-tuned version of openai/whisper-medium on the Common Voice 16.0 dataset. It achieves the following results on the evaluation set:
- Loss: 1.0525
- Wer: 100.1779
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: 500
- training_steps: 5000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.182 | 3.05 | 1000 | 0.9951 | 100.5780 |
| 0.0087 | 6.1 | 2000 | 1.0259 | 100.0889 |
| 0.004 | 9.15 | 3000 | 1.0234 | 100.0445 |
| 0.0002 | 12.2 | 4000 | 1.0484 | 100.1334 |
| 0.0002 | 15.24 | 5000 | 1.0525 | 100.1779 |
Framework versions
- Transformers 4.37.1
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.1