metadata
license: apache-2.0
base_model: facebook/wav2vec2-xls-r-300m
tags:
- generated_from_trainer
datasets:
- common_voice_13_0
metrics:
- wer
model-index:
- name: wav2vec2-large-xls-r-300m-breton-colab
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: common_voice_13_0
type: common_voice_13_0
config: br
split: test
args: br
metrics:
- name: Wer
type: wer
value: 0.508994708994709
wav2vec2-large-xls-r-300m-breton-colab
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the common_voice_13_0 dataset. It achieves the following results on the evaluation set:
- Loss: 1.2344
- Wer: 0.5090
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: 0.0003
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- 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: 500
- num_epochs: 35
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 2.8178 | 3.36 | 1000 | 1.0244 | 0.7207 |
| 0.5674 | 6.72 | 2000 | 0.9848 | 0.6341 |
| 0.33 | 10.08 | 3000 | 1.0254 | 0.6014 |
| 0.2362 | 13.45 | 4000 | 1.1387 | 0.5848 |
| 0.1777 | 16.81 | 5000 | 1.2125 | 0.5783 |
| 0.1429 | 20.17 | 6000 | 1.1952 | 0.5572 |
| 0.1076 | 23.53 | 7000 | 1.2492 | 0.5628 |
| 0.0842 | 26.89 | 8000 | 1.2103 | 0.5410 |
| 0.0666 | 30.25 | 9000 | 1.2032 | 0.5128 |
| 0.051 | 33.61 | 10000 | 1.2344 | 0.5090 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.0.1+cu118
- Datasets 2.14.4
- Tokenizers 0.13.3