Commit
•
2f164ad
1
Parent(s):
f9b1ce2
update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,70 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
datasets:
|
6 |
+
- common_voice
|
7 |
+
model-index:
|
8 |
+
- name: wav2vec2-xlsr-53-rm-vallader-with-lm
|
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-xlsr-53-rm-vallader-with-lm
|
16 |
+
|
17 |
+
This model is a fine-tuned version of [anuragshas/wav2vec2-large-xlsr-53-rm-vallader](https://huggingface.co/anuragshas/wav2vec2-large-xlsr-53-rm-vallader) on the common_voice dataset.
|
18 |
+
It achieves the following results on the evaluation set:
|
19 |
+
- Loss: 0.4552
|
20 |
+
- Wer: 0.3206
|
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: 7.5e-05
|
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_ratio: 0.112
|
48 |
+
- num_epochs: 30
|
49 |
+
|
50 |
+
### Training results
|
51 |
+
|
52 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
53 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
54 |
+
| 0.2379 | 3.12 | 100 | 0.4041 | 0.3396 |
|
55 |
+
| 0.103 | 6.25 | 200 | 0.4400 | 0.3337 |
|
56 |
+
| 0.0664 | 9.38 | 300 | 0.4239 | 0.3315 |
|
57 |
+
| 0.0578 | 12.5 | 400 | 0.4303 | 0.3267 |
|
58 |
+
| 0.0446 | 15.62 | 500 | 0.4575 | 0.3274 |
|
59 |
+
| 0.041 | 18.75 | 600 | 0.4451 | 0.3223 |
|
60 |
+
| 0.0402 | 21.88 | 700 | 0.4507 | 0.3206 |
|
61 |
+
| 0.0374 | 25.0 | 800 | 0.4649 | 0.3208 |
|
62 |
+
| 0.0371 | 28.12 | 900 | 0.4552 | 0.3206 |
|
63 |
+
|
64 |
+
|
65 |
+
### Framework versions
|
66 |
+
|
67 |
+
- Transformers 4.15.0
|
68 |
+
- Pytorch 1.10.0+cu111
|
69 |
+
- Datasets 1.18.1
|
70 |
+
- Tokenizers 0.10.3
|