Commit
•
89cc42e
1
Parent(s):
42eb712
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,94 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
metrics:
|
| 6 |
+
- wer
|
| 7 |
+
model-index:
|
| 8 |
+
- name: wav2vec2-Irish-common-voice-Fleurs-living-audio-300m
|
| 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-Irish-common-voice-Fleurs-living-audio-300m
|
| 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 None dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.3362
|
| 20 |
+
- Wer: 0.1978
|
| 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: 8
|
| 41 |
+
- eval_batch_size: 8
|
| 42 |
+
- seed: 42
|
| 43 |
+
- gradient_accumulation_steps: 3
|
| 44 |
+
- total_train_batch_size: 24
|
| 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: 18.0
|
| 49 |
+
- mixed_precision_training: Native AMP
|
| 50 |
+
|
| 51 |
+
### Training results
|
| 52 |
+
|
| 53 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 54 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
| 55 |
+
| No log | 0.56 | 200 | 2.8832 | 1.0 |
|
| 56 |
+
| No log | 1.11 | 400 | 1.1705 | 0.7788 |
|
| 57 |
+
| 3.3987 | 1.67 | 600 | 0.7739 | 0.5895 |
|
| 58 |
+
| 3.3987 | 2.23 | 800 | 0.6045 | 0.4902 |
|
| 59 |
+
| 0.8313 | 2.78 | 1000 | 0.5235 | 0.4394 |
|
| 60 |
+
| 0.8313 | 3.34 | 1200 | 0.4824 | 0.4002 |
|
| 61 |
+
| 0.8313 | 3.9 | 1400 | 0.4378 | 0.3754 |
|
| 62 |
+
| 0.5342 | 4.46 | 1600 | 0.4433 | 0.3634 |
|
| 63 |
+
| 0.5342 | 5.01 | 1800 | 0.4103 | 0.3485 |
|
| 64 |
+
| 0.3792 | 5.57 | 2000 | 0.3816 | 0.3310 |
|
| 65 |
+
| 0.3792 | 6.13 | 2200 | 0.3953 | 0.3225 |
|
| 66 |
+
| 0.3792 | 6.68 | 2400 | 0.3995 | 0.3132 |
|
| 67 |
+
| 0.2924 | 7.24 | 2600 | 0.3907 | 0.2930 |
|
| 68 |
+
| 0.2924 | 7.8 | 2800 | 0.3517 | 0.2740 |
|
| 69 |
+
| 0.2217 | 8.36 | 3000 | 0.3361 | 0.2591 |
|
| 70 |
+
| 0.2217 | 8.91 | 3200 | 0.3340 | 0.2451 |
|
| 71 |
+
| 0.2217 | 9.47 | 3400 | 0.3126 | 0.2448 |
|
| 72 |
+
| 0.1714 | 10.03 | 3600 | 0.3441 | 0.2556 |
|
| 73 |
+
| 0.1714 | 10.58 | 3800 | 0.3404 | 0.2521 |
|
| 74 |
+
| 0.1395 | 11.14 | 4000 | 0.3728 | 0.2518 |
|
| 75 |
+
| 0.1395 | 11.7 | 4200 | 0.3829 | 0.2396 |
|
| 76 |
+
| 0.1395 | 12.26 | 4400 | 0.3466 | 0.2361 |
|
| 77 |
+
| 0.1069 | 12.81 | 4600 | 0.3188 | 0.2241 |
|
| 78 |
+
| 0.1069 | 13.37 | 4800 | 0.3396 | 0.2197 |
|
| 79 |
+
| 0.0845 | 13.93 | 5000 | 0.3365 | 0.2206 |
|
| 80 |
+
| 0.0845 | 14.48 | 5200 | 0.3459 | 0.2209 |
|
| 81 |
+
| 0.0845 | 15.04 | 5400 | 0.3429 | 0.2194 |
|
| 82 |
+
| 0.0675 | 15.6 | 5600 | 0.3434 | 0.2182 |
|
| 83 |
+
| 0.0675 | 16.16 | 5800 | 0.3434 | 0.2083 |
|
| 84 |
+
| 0.0561 | 16.71 | 6000 | 0.3375 | 0.2036 |
|
| 85 |
+
| 0.0561 | 17.27 | 6200 | 0.3446 | 0.1987 |
|
| 86 |
+
| 0.0561 | 17.83 | 6400 | 0.3362 | 0.1978 |
|
| 87 |
+
|
| 88 |
+
|
| 89 |
+
### Framework versions
|
| 90 |
+
|
| 91 |
+
- Transformers 4.24.0
|
| 92 |
+
- Pytorch 1.13.0+cu117
|
| 93 |
+
- Datasets 2.7.1
|
| 94 |
+
- Tokenizers 0.13.2
|