metadata
library_name: transformers
license: apache-2.0
base_model: facebook/wav2vec2-large-xlsr-53-portuguese
tags:
- generated_from_trainer
datasets:
- common_voice_17_0
metrics:
- wer
model-index:
- name: xlsr53-ptbr-2
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: common_voice_17_0
type: common_voice_17_0
config: pt
split: None
args: pt
metrics:
- name: Wer
type: wer
value: 0.9949969678593087
xlsr53-ptbr-2
This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53-portuguese on the common_voice_17_0 dataset. It achieves the following results on the evaluation set:
- Loss: 3.9879
- Wer: 0.9950
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-06
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 12.583 | 1.0 | 1250 | 8.3731 | 0.9930 |
| 8.4082 | 2.0 | 2500 | 6.7542 | 0.9977 |
| 9.7515 | 3.0 | 3750 | 6.4559 | 1.0044 |
| 10.5866 | 4.0 | 5000 | 6.1102 | 1.0047 |
| 9.6095 | 5.0 | 6250 | 5.5412 | 1.0012 |
| 7.9527 | 6.0 | 7500 | 4.7235 | 0.9961 |
| 6.1536 | 7.0 | 8750 | 4.2559 | 0.9971 |
| 6.3449 | 8.0 | 10000 | 4.1116 | 0.9956 |
| 5.4191 | 9.0 | 11250 | 4.0244 | 0.9955 |
| 4.8775 | 10.0 | 12500 | 3.9879 | 0.9951 |
Framework versions
- Transformers 4.47.1
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.21.0