metadata
language:
- hi
license: apache-2.0
tags:
- hf-asr-leaderboard
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_11_0
metrics:
- wer
base_model: openai/whisper-large-v3
model-index:
- name: whisper-small-dataset
results:
- task:
type: automatic-speech-recognition
name: Automatic Speech Recognition
dataset:
name: Common Voice 11.0
type: mozilla-foundation/common_voice_11_0
config: hi
split: None
args: 'config: hi, split: test'
metrics:
- type: wer
value: 48.5207100591716
name: Wer
whisper-small-dataset
This model is a fine-tuned version of openai/whisper-large-v3 on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.2599
- Wer: 48.5207
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: 4
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- 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: 5
- training_steps: 40
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| No log | 1.6 | 10 | 0.3733 | 50.2959 |
| No log | 3.2 | 20 | 0.2663 | 52.0710 |
| 0.2997 | 4.8 | 30 | 0.2667 | 48.5207 |
| 0.2997 | 6.4 | 40 | 0.2599 | 48.5207 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.2.0+cu121
- Datasets 2.17.0
- Tokenizers 0.15.2