File size: 2,038 Bytes
41a5f41 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 |
---
language:
- pt
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
datasets:
- google/fleurs
metrics:
- wer
model-index:
- name: Whisper small pt - m1
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: fleurs
type: google/fleurs
config: pt_br
split: None
args: 'config: pt_br, split: test, train'
metrics:
- name: Wer
type: wer
value: 8.993465281369
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# Whisper small pt - m1
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the fleurs dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2815
- Wer: 8.9935
## 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: 2500
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|:-------------:|:-------:|:----:|:---------------:|:------:|
| 0.0892 | 2.8571 | 500 | 0.2241 | 8.9747 |
| 0.0041 | 5.7143 | 1000 | 0.2586 | 8.9653 |
| 0.0014 | 8.5714 | 1500 | 0.2702 | 8.8665 |
| 0.001 | 11.4286 | 2000 | 0.2788 | 8.9512 |
| 0.0008 | 14.2857 | 2500 | 0.2815 | 8.9935 |
### Framework versions
- Transformers 4.42.4
- Pytorch 2.3.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1
|