Commit
•
865c8fa
1
Parent(s):
aeedc61
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,86 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- sv-SE
|
| 4 |
+
license: apache-2.0
|
| 5 |
+
tags:
|
| 6 |
+
- automatic-speech-recognition
|
| 7 |
+
- common_voice
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
datasets:
|
| 10 |
+
- common_voice
|
| 11 |
+
model-index:
|
| 12 |
+
- name: wav2vec2-common_voice-tr-demo
|
| 13 |
+
results: []
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 17 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 18 |
+
|
| 19 |
+
# wav2vec2-common_voice-tr-demo
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the COMMON_VOICE - SV-SE dataset.
|
| 22 |
+
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 0.5528
|
| 24 |
+
- Wer: 0.3811
|
| 25 |
+
|
| 26 |
+
## Model description
|
| 27 |
+
|
| 28 |
+
More information needed
|
| 29 |
+
|
| 30 |
+
## Intended uses & limitations
|
| 31 |
+
|
| 32 |
+
More information needed
|
| 33 |
+
|
| 34 |
+
## Training and evaluation data
|
| 35 |
+
|
| 36 |
+
More information needed
|
| 37 |
+
|
| 38 |
+
## Training procedure
|
| 39 |
+
|
| 40 |
+
### Training hyperparameters
|
| 41 |
+
|
| 42 |
+
The following hyperparameters were used during training:
|
| 43 |
+
- learning_rate: 0.0003
|
| 44 |
+
- train_batch_size: 16
|
| 45 |
+
- eval_batch_size: 8
|
| 46 |
+
- seed: 42
|
| 47 |
+
- gradient_accumulation_steps: 2
|
| 48 |
+
- total_train_batch_size: 32
|
| 49 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 50 |
+
- lr_scheduler_type: linear
|
| 51 |
+
- lr_scheduler_warmup_steps: 500
|
| 52 |
+
- num_epochs: 15.0
|
| 53 |
+
- mixed_precision_training: Native AMP
|
| 54 |
+
|
| 55 |
+
### Training results
|
| 56 |
+
|
| 57 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 58 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
| 59 |
+
| No log | 0.74 | 100 | 3.4444 | 1.0 |
|
| 60 |
+
| No log | 1.47 | 200 | 2.9421 | 1.0 |
|
| 61 |
+
| No log | 2.21 | 300 | 2.2802 | 1.0137 |
|
| 62 |
+
| No log | 2.94 | 400 | 0.9683 | 0.7611 |
|
| 63 |
+
| 3.7264 | 3.68 | 500 | 0.7941 | 0.6594 |
|
| 64 |
+
| 3.7264 | 4.41 | 600 | 0.6695 | 0.5751 |
|
| 65 |
+
| 3.7264 | 5.15 | 700 | 0.6507 | 0.5314 |
|
| 66 |
+
| 3.7264 | 5.88 | 800 | 0.5731 | 0.4927 |
|
| 67 |
+
| 3.7264 | 6.62 | 900 | 0.5723 | 0.4580 |
|
| 68 |
+
| 0.4592 | 7.35 | 1000 | 0.5913 | 0.4479 |
|
| 69 |
+
| 0.4592 | 8.09 | 1100 | 0.5562 | 0.4423 |
|
| 70 |
+
| 0.4592 | 8.82 | 1200 | 0.5566 | 0.4292 |
|
| 71 |
+
| 0.4592 | 9.56 | 1300 | 0.5492 | 0.4303 |
|
| 72 |
+
| 0.4592 | 10.29 | 1400 | 0.5665 | 0.4331 |
|
| 73 |
+
| 0.2121 | 11.03 | 1500 | 0.5610 | 0.4084 |
|
| 74 |
+
| 0.2121 | 11.76 | 1600 | 0.5703 | 0.4014 |
|
| 75 |
+
| 0.2121 | 12.5 | 1700 | 0.5669 | 0.3898 |
|
| 76 |
+
| 0.2121 | 13.24 | 1800 | 0.5586 | 0.3962 |
|
| 77 |
+
| 0.2121 | 13.97 | 1900 | 0.5656 | 0.3897 |
|
| 78 |
+
| 0.1326 | 14.71 | 2000 | 0.5565 | 0.3813 |
|
| 79 |
+
|
| 80 |
+
|
| 81 |
+
### Framework versions
|
| 82 |
+
|
| 83 |
+
- Transformers 4.16.0.dev0
|
| 84 |
+
- Pytorch 1.10.1+cu113
|
| 85 |
+
- Datasets 1.18.0
|
| 86 |
+
- Tokenizers 0.10.3
|