Commit
•
2237dbe
1
Parent(s):
36cba1f
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
datasets:
|
| 6 |
+
- common_voice
|
| 7 |
+
model-index:
|
| 8 |
+
- name: wav2vec2-base-checkpoint-7.1
|
| 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-base-checkpoint-7.1
|
| 16 |
+
|
| 17 |
+
This model is a fine-tuned version of [jiobiala24/wav2vec2-base-checkpoint-6](https://huggingface.co/jiobiala24/wav2vec2-base-checkpoint-6) on the common_voice dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.9369
|
| 20 |
+
- Wer: 0.3243
|
| 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: 0.0001
|
| 40 |
+
- train_batch_size: 32
|
| 41 |
+
- eval_batch_size: 8
|
| 42 |
+
- seed: 42
|
| 43 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 44 |
+
- lr_scheduler_type: linear
|
| 45 |
+
- lr_scheduler_warmup_steps: 1000
|
| 46 |
+
- num_epochs: 30
|
| 47 |
+
- mixed_precision_training: Native AMP
|
| 48 |
+
|
| 49 |
+
### Training results
|
| 50 |
+
|
| 51 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 52 |
+
|:-------------:|:-----:|:-----:|:---------------:|:------:|
|
| 53 |
+
| 0.3124 | 1.75 | 1000 | 0.5602 | 0.3403 |
|
| 54 |
+
| 0.2428 | 3.5 | 2000 | 0.5924 | 0.3431 |
|
| 55 |
+
| 0.1884 | 5.24 | 3000 | 0.6161 | 0.3423 |
|
| 56 |
+
| 0.1557 | 6.99 | 4000 | 0.6570 | 0.3415 |
|
| 57 |
+
| 0.1298 | 8.74 | 5000 | 0.6837 | 0.3446 |
|
| 58 |
+
| 0.1141 | 10.49 | 6000 | 0.7304 | 0.3396 |
|
| 59 |
+
| 0.1031 | 12.24 | 7000 | 0.7264 | 0.3410 |
|
| 60 |
+
| 0.0916 | 13.99 | 8000 | 0.7229 | 0.3387 |
|
| 61 |
+
| 0.0835 | 15.73 | 9000 | 0.8078 | 0.3458 |
|
| 62 |
+
| 0.0761 | 17.48 | 10000 | 0.8304 | 0.3408 |
|
| 63 |
+
| 0.0693 | 19.23 | 11000 | 0.8290 | 0.3387 |
|
| 64 |
+
| 0.0646 | 20.98 | 12000 | 0.8593 | 0.3372 |
|
| 65 |
+
| 0.0605 | 22.73 | 13000 | 0.8728 | 0.3345 |
|
| 66 |
+
| 0.0576 | 24.48 | 14000 | 0.9111 | 0.3297 |
|
| 67 |
+
| 0.0529 | 26.22 | 15000 | 0.9247 | 0.3273 |
|
| 68 |
+
| 0.0492 | 27.97 | 16000 | 0.9248 | 0.3250 |
|
| 69 |
+
| 0.0472 | 29.72 | 17000 | 0.9369 | 0.3243 |
|
| 70 |
+
|
| 71 |
+
|
| 72 |
+
### Framework versions
|
| 73 |
+
|
| 74 |
+
- Transformers 4.11.3
|
| 75 |
+
- Pytorch 1.10.0+cu111
|
| 76 |
+
- Datasets 1.13.3
|
| 77 |
+
- Tokenizers 0.10.3
|