|
--- |
|
license: apache-2.0 |
|
base_model: facebook/wav2vec2-xls-r-300m |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- common_voice_16_1 |
|
metrics: |
|
- wer |
|
model-index: |
|
- name: wav2vec2-turkish-300m-3 |
|
results: |
|
- task: |
|
name: Automatic Speech Recognition |
|
type: automatic-speech-recognition |
|
dataset: |
|
name: common_voice_16_1 |
|
type: common_voice_16_1 |
|
config: tr |
|
split: test |
|
args: tr |
|
metrics: |
|
- name: Wer |
|
type: wer |
|
value: 0.2453349153273649 |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# wav2vec2-turkish-300m-3 |
|
|
|
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the common_voice_16_1 dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.2968 |
|
- Wer: 0.2453 |
|
|
|
## 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: 32 |
|
- eval_batch_size: 16 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 0.1 |
|
- num_epochs: 20 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Wer | |
|
|:-------------:|:-------:|:-----:|:---------------:|:------:| |
|
| 2.015 | 0.3652 | 500 | 0.5674 | 0.5880 | |
|
| 0.5687 | 0.7305 | 1000 | 0.4913 | 0.5483 | |
|
| 0.4727 | 1.0957 | 1500 | 0.4382 | 0.4868 | |
|
| 0.3713 | 1.4609 | 2000 | 0.3941 | 0.4761 | |
|
| 0.3653 | 1.8262 | 2500 | 0.3978 | 0.4609 | |
|
| 0.3457 | 2.1914 | 3000 | 0.3570 | 0.4201 | |
|
| 0.3108 | 2.5566 | 3500 | 0.3273 | 0.4045 | |
|
| 0.2985 | 2.9218 | 4000 | 0.3559 | 0.4253 | |
|
| 0.2768 | 3.2871 | 4500 | 0.3484 | 0.4288 | |
|
| 0.2702 | 3.6523 | 5000 | 0.3422 | 0.3988 | |
|
| 0.2626 | 4.0175 | 5500 | 0.3312 | 0.3875 | |
|
| 0.246 | 4.3828 | 6000 | 0.3175 | 0.3735 | |
|
| 0.2373 | 4.7480 | 6500 | 0.3126 | 0.3750 | |
|
| 0.234 | 5.1132 | 7000 | 0.3289 | 0.3703 | |
|
| 0.2225 | 5.4785 | 7500 | 0.3170 | 0.3700 | |
|
| 0.2094 | 5.8437 | 8000 | 0.3127 | 0.3611 | |
|
| 0.1961 | 6.2089 | 8500 | 0.3130 | 0.3604 | |
|
| 0.1927 | 6.5741 | 9000 | 0.3167 | 0.3491 | |
|
| 0.1963 | 6.9394 | 9500 | 0.2983 | 0.3451 | |
|
| 0.1757 | 7.3046 | 10000 | 0.3044 | 0.3403 | |
|
| 0.1732 | 7.6698 | 10500 | 0.2988 | 0.3407 | |
|
| 0.1737 | 8.0351 | 11000 | 0.3128 | 0.3367 | |
|
| 0.1686 | 8.4003 | 11500 | 0.2954 | 0.3296 | |
|
| 0.1588 | 8.7655 | 12000 | 0.3226 | 0.3265 | |
|
| 0.1481 | 9.1308 | 12500 | 0.2946 | 0.3172 | |
|
| 0.1434 | 9.4960 | 13000 | 0.2981 | 0.3202 | |
|
| 0.146 | 9.8612 | 13500 | 0.2936 | 0.3150 | |
|
| 0.1352 | 10.2264 | 14000 | 0.2895 | 0.3091 | |
|
| 0.1304 | 10.5917 | 14500 | 0.2932 | 0.3071 | |
|
| 0.1253 | 10.9569 | 15000 | 0.2946 | 0.2997 | |
|
| 0.12 | 11.3221 | 15500 | 0.2967 | 0.3065 | |
|
| 0.1179 | 11.6874 | 16000 | 0.2856 | 0.3037 | |
|
| 0.1185 | 12.0526 | 16500 | 0.2753 | 0.2973 | |
|
| 0.1128 | 12.4178 | 17000 | 0.2954 | 0.2935 | |
|
| 0.1054 | 12.7831 | 17500 | 0.2917 | 0.2916 | |
|
| 0.1026 | 13.1483 | 18000 | 0.2878 | 0.2820 | |
|
| 0.0981 | 13.5135 | 18500 | 0.2882 | 0.2863 | |
|
| 0.0936 | 13.8787 | 19000 | 0.2758 | 0.2774 | |
|
| 0.0911 | 14.2440 | 19500 | 0.2867 | 0.2811 | |
|
| 0.0881 | 14.6092 | 20000 | 0.2952 | 0.2760 | |
|
| 0.0809 | 14.9744 | 20500 | 0.2996 | 0.2772 | |
|
| 0.0815 | 15.3397 | 21000 | 0.2806 | 0.2694 | |
|
| 0.078 | 15.7049 | 21500 | 0.3050 | 0.2717 | |
|
| 0.0727 | 16.0701 | 22000 | 0.2871 | 0.2682 | |
|
| 0.0716 | 16.4354 | 22500 | 0.2935 | 0.2667 | |
|
| 0.0672 | 16.8006 | 23000 | 0.2917 | 0.2632 | |
|
| 0.0666 | 17.1658 | 23500 | 0.3075 | 0.2584 | |
|
| 0.0654 | 17.5310 | 24000 | 0.3025 | 0.2580 | |
|
| 0.0616 | 17.8963 | 24500 | 0.2952 | 0.2550 | |
|
| 0.0609 | 18.2615 | 25000 | 0.3077 | 0.2567 | |
|
| 0.0604 | 18.6267 | 25500 | 0.3040 | 0.2513 | |
|
| 0.0549 | 18.9920 | 26000 | 0.3043 | 0.2481 | |
|
| 0.0516 | 19.3572 | 26500 | 0.3036 | 0.2476 | |
|
| 0.0543 | 19.7224 | 27000 | 0.2968 | 0.2453 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.40.0 |
|
- Pytorch 2.2.2+cu121 |
|
- Datasets 2.17.1 |
|
- Tokenizers 0.19.1 |
|
|
