metadata
language:
- gn
license: apache-2.0
base_model: openai/whisper-base
tags:
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_16_1
metrics:
- wer
model-index:
- name: Common Voice 16 - Guarani
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: Common Voice 16
type: mozilla-foundation/common_voice_16_1
config: gn
split: None
args: gn
metrics:
- name: Wer
type: wer
value: 58.291457286432156
Common Voice 16 - Guarani
This model is a fine-tuned version of openai/whisper-base on the Common Voice 16 dataset. It achieves the following results on the evaluation set:
- Loss: 0.5062
- Wer: 58.2915
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: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant_with_warmup
- lr_scheduler_warmup_steps: 3000
- training_steps: 2000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 2.9797 | 0.4955 | 500 | 1.0523 | 92.2948 |
| 1.1502 | 0.9911 | 1000 | 0.6757 | 87.0463 |
| 0.7171 | 1.4866 | 1500 | 0.5723 | 63.7074 |
| 0.5485 | 1.9822 | 2000 | 0.5062 | 58.2915 |
Framework versions
- Transformers 4.44.0
- Pytorch 2.3.1+cu121
- Datasets 2.21.0
- Tokenizers 0.19.1