update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,83 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
datasets:
|
| 6 |
+
- common_voice
|
| 7 |
+
model-index:
|
| 8 |
+
- name: wav2vec2-large-xls-r-300m-turkish-colab_common_voice-8_4
|
| 9 |
+
results: []
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 13 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 14 |
+
|
| 15 |
+
# wav2vec2-large-xls-r-300m-turkish-colab_common_voice-8_4
|
| 16 |
+
|
| 17 |
+
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 dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.3201
|
| 20 |
+
- Wer: 0.3295
|
| 21 |
+
|
| 22 |
+
## Model description
|
| 23 |
+
|
| 24 |
+
More information needed
|
| 25 |
+
|
| 26 |
+
## Intended uses & limitations
|
| 27 |
+
|
| 28 |
+
More information needed
|
| 29 |
+
|
| 30 |
+
## Training and evaluation data
|
| 31 |
+
|
| 32 |
+
More information needed
|
| 33 |
+
|
| 34 |
+
## Training procedure
|
| 35 |
+
|
| 36 |
+
### Training hyperparameters
|
| 37 |
+
|
| 38 |
+
The following hyperparameters were used during training:
|
| 39 |
+
- learning_rate: 0.0003
|
| 40 |
+
- train_batch_size: 16
|
| 41 |
+
- eval_batch_size: 8
|
| 42 |
+
- seed: 42
|
| 43 |
+
- gradient_accumulation_steps: 2
|
| 44 |
+
- total_train_batch_size: 32
|
| 45 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 46 |
+
- lr_scheduler_type: linear
|
| 47 |
+
- lr_scheduler_warmup_steps: 500
|
| 48 |
+
- num_epochs: 11
|
| 49 |
+
- mixed_precision_training: Native AMP
|
| 50 |
+
|
| 51 |
+
### Training results
|
| 52 |
+
|
| 53 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 54 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
| 55 |
+
| 5.9268 | 0.51 | 400 | 1.3204 | 0.9175 |
|
| 56 |
+
| 0.7491 | 1.02 | 800 | 0.5880 | 0.6388 |
|
| 57 |
+
| 0.4911 | 1.53 | 1200 | 0.4680 | 0.5613 |
|
| 58 |
+
| 0.4265 | 2.04 | 1600 | 0.4213 | 0.5059 |
|
| 59 |
+
| 0.3473 | 2.55 | 2000 | 0.4199 | 0.4955 |
|
| 60 |
+
| 0.3291 | 3.07 | 2400 | 0.4323 | 0.5061 |
|
| 61 |
+
| 0.2819 | 3.58 | 2800 | 0.4026 | 0.4490 |
|
| 62 |
+
| 0.2628 | 4.09 | 3200 | 0.3831 | 0.4446 |
|
| 63 |
+
| 0.2371 | 4.6 | 3600 | 0.3622 | 0.4234 |
|
| 64 |
+
| 0.2274 | 5.11 | 4000 | 0.3473 | 0.4012 |
|
| 65 |
+
| 0.2051 | 5.62 | 4400 | 0.3471 | 0.3998 |
|
| 66 |
+
| 0.1985 | 6.13 | 4800 | 0.3759 | 0.4088 |
|
| 67 |
+
| 0.1767 | 6.64 | 5200 | 0.3620 | 0.4012 |
|
| 68 |
+
| 0.1707 | 7.15 | 5600 | 0.3415 | 0.3700 |
|
| 69 |
+
| 0.1559 | 7.66 | 6000 | 0.3317 | 0.3661 |
|
| 70 |
+
| 0.147 | 8.17 | 6400 | 0.3265 | 0.3618 |
|
| 71 |
+
| 0.1339 | 8.68 | 6800 | 0.3293 | 0.3586 |
|
| 72 |
+
| 0.126 | 9.2 | 7200 | 0.3386 | 0.3458 |
|
| 73 |
+
| 0.1149 | 9.71 | 7600 | 0.3305 | 0.3397 |
|
| 74 |
+
| 0.1051 | 10.22 | 8000 | 0.3235 | 0.3354 |
|
| 75 |
+
| 0.1005 | 10.73 | 8400 | 0.3201 | 0.3295 |
|
| 76 |
+
|
| 77 |
+
|
| 78 |
+
### Framework versions
|
| 79 |
+
|
| 80 |
+
- Transformers 4.11.3
|
| 81 |
+
- Pytorch 1.10.0+cu113
|
| 82 |
+
- Datasets 2.1.0
|
| 83 |
+
- Tokenizers 0.10.3
|