metadata
language:
- tr
tags:
- automatic-speech-recognition
- common_voice
- generated_from_trainer
datasets:
- common_voice
model-index:
- name: ''
results: []
This model is a fine-tuned version of hf-test/xls-r-dummy on the COMMON_VOICE - TR dataset. It achieves the following results on the evaluation set:
- Loss: 59.6880
- Wer: 1.0
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: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 5.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| No log | 0.92 | 100 | 149.2460 | 1.0032 |
| No log | 1.83 | 200 | 135.3586 | 1.0 |
| No log | 2.75 | 300 | 114.3390 | 1.0 |
| No log | 3.67 | 400 | 88.3077 | 1.0 |
| 67.8005 | 4.59 | 500 | 64.6167 | 1.0 |
Framework versions
- Transformers 4.17.0.dev0
- Pytorch 1.10.2+cu102
- Datasets 1.18.2.dev0
- Tokenizers 0.11.0