Commit
·
b692d6d
1
Parent(s):
55c11c1
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
datasets:
|
| 6 |
+
- common_voice
|
| 7 |
+
model-index:
|
| 8 |
+
- name: wav2vec2-large-xlsr-tamil-commonvoice
|
| 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-xlsr-tamil-commonvoice
|
| 16 |
+
|
| 17 |
+
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 dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.6145
|
| 20 |
+
- Wer: 0.8512
|
| 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: 200
|
| 48 |
+
- num_epochs: 20
|
| 49 |
+
- mixed_precision_training: Native AMP
|
| 50 |
+
|
| 51 |
+
### Training results
|
| 52 |
+
|
| 53 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 54 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
| 55 |
+
| 12.0478 | 1.05 | 100 | 3.3867 | 1.0 |
|
| 56 |
+
| 3.2522 | 2.11 | 200 | 3.2770 | 1.0 |
|
| 57 |
+
| 3.1689 | 3.16 | 300 | 3.1135 | 1.0039 |
|
| 58 |
+
| 2.9278 | 4.21 | 400 | 2.0485 | 1.3109 |
|
| 59 |
+
| 1.3592 | 5.26 | 500 | 0.8044 | 1.0988 |
|
| 60 |
+
| 0.7472 | 6.32 | 600 | 0.6571 | 0.9474 |
|
| 61 |
+
| 0.5842 | 7.37 | 700 | 0.6079 | 0.9477 |
|
| 62 |
+
| 0.4831 | 8.42 | 800 | 0.6083 | 0.9491 |
|
| 63 |
+
| 0.4259 | 9.47 | 900 | 0.5916 | 0.8973 |
|
| 64 |
+
| 0.3817 | 10.53 | 1000 | 0.6070 | 0.9147 |
|
| 65 |
+
| 0.338 | 11.58 | 1100 | 0.5873 | 0.8617 |
|
| 66 |
+
| 0.3123 | 12.63 | 1200 | 0.5983 | 0.8844 |
|
| 67 |
+
| 0.287 | 13.68 | 1300 | 0.6146 | 0.8988 |
|
| 68 |
+
| 0.2706 | 14.74 | 1400 | 0.6068 | 0.8754 |
|
| 69 |
+
| 0.2505 | 15.79 | 1500 | 0.5996 | 0.8638 |
|
| 70 |
+
| 0.2412 | 16.84 | 1600 | 0.6106 | 0.8481 |
|
| 71 |
+
| 0.2176 | 17.89 | 1700 | 0.6152 | 0.8520 |
|
| 72 |
+
| 0.2255 | 18.95 | 1800 | 0.6150 | 0.8540 |
|
| 73 |
+
| 0.216 | 20.0 | 1900 | 0.6145 | 0.8512 |
|
| 74 |
+
|
| 75 |
+
|
| 76 |
+
### Framework versions
|
| 77 |
+
|
| 78 |
+
- Transformers 4.11.3
|
| 79 |
+
- Pytorch 1.10.0+cu102
|
| 80 |
+
- Datasets 1.13.3
|
| 81 |
+
- Tokenizers 0.10.3
|