metadata
library_name: transformers
license: apache-2.0
base_model: facebook/wav2vec2-base
tags:
- audio-classification
- generated_from_trainer
datasets:
- common_language
metrics:
- accuracy
model-index:
- name: wav2vec2-base-lang-id
results:
- task:
name: Audio Classification
type: audio-classification
dataset:
name: common_language
type: common_language
config: full
split: validation
args: full
metrics:
- name: Accuracy
type: accuracy
value: 0.7800611413043478
wav2vec2-base-lang-id
This model is a fine-tuned version of facebook/wav2vec2-base on the common_language dataset. It achieves the following results on the evaluation set:
- Loss: 1.2554
- Accuracy: 0.7801
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: 1
- seed: 0
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- 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
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.58 | 0.9989 | 693 | 2.5609 | 0.2899 |
| 1.8581 | 1.9989 | 1386 | 2.1486 | 0.4008 |
| 1.3784 | 2.9989 | 2079 | 1.5906 | 0.5666 |
| 0.976 | 3.9989 | 2772 | 1.4036 | 0.6318 |
| 0.6109 | 4.9989 | 3465 | 1.3022 | 0.6695 |
| 0.4357 | 5.9989 | 4158 | 1.2386 | 0.7138 |
| 0.23 | 6.9989 | 4851 | 1.3078 | 0.7221 |
| 0.1461 | 7.9989 | 5544 | 1.2247 | 0.7534 |
| 0.0567 | 8.9989 | 6237 | 1.3279 | 0.7646 |
| 0.0375 | 9.9989 | 6930 | 1.2554 | 0.7801 |
Framework versions
- Transformers 4.49.0.dev0
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0