metadata
base_model: facebook/wav2vec2-xls-r-300m
datasets:
- common_voice_16_1
license: apache-2.0
metrics:
- wer
tags:
- generated_from_trainer
model-index:
- name: wav2vec2-large-xls-r-300m-amharic-demo-colab
results:
- task:
type: automatic-speech-recognition
name: Automatic Speech Recognition
dataset:
name: common_voice_16_1
type: common_voice_16_1
config: am
split: test
args: am
metrics:
- type: wer
value: 0.8992661774516344
name: Wer
wav2vec2-large-xls-r-300m-amharic-demo-colab
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the common_voice_16_1 dataset. It achieves the following results on the evaluation set:
- Loss: 1.6489
- Wer: 0.8993
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: 100
- num_epochs: 30
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 12.8229 | 2.5 | 100 | 4.1682 | 1.0 |
| 4.1232 | 5.0 | 200 | 4.0821 | 1.0 |
| 4.0475 | 7.5 | 300 | 4.0087 | 1.0 |
| 3.9841 | 10.0 | 400 | 3.9677 | 1.0 |
| 3.9469 | 12.5 | 500 | 3.9503 | 1.0 |
| 3.7544 | 15.0 | 600 | 3.3452 | 1.0 |
| 2.1016 | 17.5 | 700 | 1.8871 | 0.9800 |
| 0.9969 | 20.0 | 800 | 1.7061 | 0.9813 |
| 0.6112 | 22.5 | 900 | 1.6420 | 0.9513 |
| 0.4384 | 25.0 | 1000 | 1.6287 | 0.9466 |
| 0.3355 | 27.5 | 1100 | 1.6593 | 0.9273 |
| 0.293 | 30.0 | 1200 | 1.6489 | 0.8993 |
Framework versions
- Transformers 4.42.0
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1