will commited on
Commit ·
5ba8c7c
1
Parent(s): bb53115
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
model-index:
|
| 6 |
+
- name: wav2vec2-large-robust-ft-timit
|
| 7 |
+
results: []
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 11 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 12 |
+
|
| 13 |
+
# wav2vec2-large-robust-ft-timit
|
| 14 |
+
|
| 15 |
+
This model is a fine-tuned version of [facebook/wav2vec2-large-robust](https://huggingface.co/facebook/wav2vec2-large-robust) on the None dataset.
|
| 16 |
+
It achieves the following results on the evaluation set:
|
| 17 |
+
- Loss: 0.2768
|
| 18 |
+
- Wer: 0.2321
|
| 19 |
+
|
| 20 |
+
## Model description
|
| 21 |
+
|
| 22 |
+
More information needed
|
| 23 |
+
|
| 24 |
+
## Intended uses & limitations
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Training and evaluation data
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training procedure
|
| 33 |
+
|
| 34 |
+
### Training hyperparameters
|
| 35 |
+
|
| 36 |
+
The following hyperparameters were used during training:
|
| 37 |
+
- learning_rate: 0.0001
|
| 38 |
+
- train_batch_size: 8
|
| 39 |
+
- eval_batch_size: 8
|
| 40 |
+
- seed: 42
|
| 41 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 42 |
+
- lr_scheduler_type: linear
|
| 43 |
+
- lr_scheduler_warmup_steps: 1000
|
| 44 |
+
- num_epochs: 30
|
| 45 |
+
- mixed_precision_training: Native AMP
|
| 46 |
+
|
| 47 |
+
### Training results
|
| 48 |
+
|
| 49 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 50 |
+
|:-------------:|:-----:|:-----:|:---------------:|:------:|
|
| 51 |
+
| 6.6175 | 1.0 | 500 | 3.3025 | 1.0 |
|
| 52 |
+
| 3.0746 | 2.01 | 1000 | 2.9598 | 1.0 |
|
| 53 |
+
| 1.967 | 3.01 | 1500 | 0.6760 | 0.5607 |
|
| 54 |
+
| 0.7545 | 4.02 | 2000 | 0.4500 | 0.4567 |
|
| 55 |
+
| 0.5415 | 5.02 | 2500 | 0.3702 | 0.3882 |
|
| 56 |
+
| 0.4445 | 6.02 | 3000 | 0.3421 | 0.3584 |
|
| 57 |
+
| 0.3601 | 7.03 | 3500 | 0.2947 | 0.3096 |
|
| 58 |
+
| 0.3098 | 8.03 | 4000 | 0.2740 | 0.2894 |
|
| 59 |
+
| 0.2606 | 9.04 | 4500 | 0.2725 | 0.2787 |
|
| 60 |
+
| 0.238 | 10.04 | 5000 | 0.2549 | 0.2617 |
|
| 61 |
+
| 0.2142 | 11.04 | 5500 | 0.2485 | 0.2530 |
|
| 62 |
+
| 0.1787 | 12.05 | 6000 | 0.2683 | 0.2514 |
|
| 63 |
+
| 0.1652 | 13.05 | 6500 | 0.2559 | 0.2476 |
|
| 64 |
+
| 0.1569 | 14.06 | 7000 | 0.2777 | 0.2470 |
|
| 65 |
+
| 0.1443 | 15.06 | 7500 | 0.2661 | 0.2431 |
|
| 66 |
+
| 0.1335 | 16.06 | 8000 | 0.2717 | 0.2422 |
|
| 67 |
+
| 0.1291 | 17.07 | 8500 | 0.2672 | 0.2428 |
|
| 68 |
+
| 0.1192 | 18.07 | 9000 | 0.2684 | 0.2395 |
|
| 69 |
+
| 0.1144 | 19.08 | 9500 | 0.2770 | 0.2411 |
|
| 70 |
+
| 0.1052 | 20.08 | 10000 | 0.2831 | 0.2379 |
|
| 71 |
+
| 0.1004 | 21.08 | 10500 | 0.2847 | 0.2375 |
|
| 72 |
+
| 0.1053 | 22.09 | 11000 | 0.2851 | 0.2360 |
|
| 73 |
+
| 0.1005 | 23.09 | 11500 | 0.2807 | 0.2361 |
|
| 74 |
+
| 0.0904 | 24.1 | 12000 | 0.2764 | 0.2346 |
|
| 75 |
+
| 0.0876 | 25.1 | 12500 | 0.2774 | 0.2325 |
|
| 76 |
+
| 0.0883 | 26.1 | 13000 | 0.2768 | 0.2313 |
|
| 77 |
+
| 0.0848 | 27.11 | 13500 | 0.2840 | 0.2307 |
|
| 78 |
+
| 0.0822 | 28.11 | 14000 | 0.2812 | 0.2316 |
|
| 79 |
+
| 0.09 | 29.12 | 14500 | 0.2768 | 0.2321 |
|
| 80 |
+
|
| 81 |
+
|
| 82 |
+
### Framework versions
|
| 83 |
+
|
| 84 |
+
- Transformers 4.19.2
|
| 85 |
+
- Pytorch 1.8.2+cu111
|
| 86 |
+
- Datasets 1.17.0
|
| 87 |
+
- Tokenizers 0.11.6
|