Commit
•
06f71f9
1
Parent(s):
47bb259
update model card README.md
Browse files
README.md
CHANGED
@@ -1,57 +1,12 @@
|
|
1 |
---
|
2 |
-
|
3 |
-
- pa-IN
|
4 |
-
|
5 |
-
license: apache-2.0
|
6 |
tags:
|
7 |
-
-
|
8 |
-
- robust-speech-event
|
9 |
datasets:
|
10 |
-
-
|
11 |
-
metrics:
|
12 |
-
- wer
|
13 |
-
- cer
|
14 |
model-index:
|
15 |
- name: wav2vec2-large-xlsr-53-punjabi
|
16 |
-
results:
|
17 |
-
- task:
|
18 |
-
type: automatic-speech-recognition # Required. Example: automatic-speech-recognition
|
19 |
-
name: Speech Recognition # Optional. Example: Speech Recognition
|
20 |
-
dataset:
|
21 |
-
type: mozilla-foundation/common_voice_7_0 # Required. Example: common_voice. Use dataset id from https://hf.co/datasets
|
22 |
-
name: Common Voice pa-IN # Required. Example: Common Voice zh-CN
|
23 |
-
args: pa-IN # Optional. Example: zh-CN
|
24 |
-
metrics:
|
25 |
-
- type: wer # Required. Example: wer
|
26 |
-
value: 39.42 # Required. Example: 20.90
|
27 |
-
name: Test WER # Optional. Example: Test WER
|
28 |
-
args:
|
29 |
-
- learning_rate: 0.0003
|
30 |
-
- train_batch_size: 16
|
31 |
-
- eval_batch_size: 8
|
32 |
-
- seed: 42
|
33 |
-
- gradient_accumulation_steps: 2
|
34 |
-
- total_train_batch_size: 32
|
35 |
-
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
36 |
-
- lr_scheduler_type: linear
|
37 |
-
- lr_scheduler_warmup_steps: 200
|
38 |
-
- num_epochs: 30
|
39 |
-
- mixed_precision_training: Native AMP # Optional. Example for BLEU: max_order
|
40 |
-
- type: cer # Required. Example: wer
|
41 |
-
value: 12.99 # Required. Example: 20.90
|
42 |
-
name: Test CER # Optional. Example: Test WER
|
43 |
-
args:
|
44 |
-
- learning_rate: 0.0003
|
45 |
-
- train_batch_size: 16
|
46 |
-
- eval_batch_size: 8
|
47 |
-
- seed: 42
|
48 |
-
- gradient_accumulation_steps: 2
|
49 |
-
- total_train_batch_size: 32
|
50 |
-
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
51 |
-
- lr_scheduler_type: linear
|
52 |
-
- lr_scheduler_warmup_steps: 200
|
53 |
-
- num_epochs: 30
|
54 |
-
- mixed_precision_training: Native AMP # Optional. Example for BLEU: max_order
|
55 |
---
|
56 |
|
57 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -59,11 +14,23 @@ should probably proofread and complete it, then remove this comment. -->
|
|
59 |
|
60 |
# wav2vec2-large-xlsr-53-punjabi
|
61 |
|
62 |
-
This model is a fine-tuned version of [
|
63 |
It achieves the following results on the evaluation set:
|
64 |
-
- Loss:
|
65 |
-
- Wer: 0.
|
66 |
-
- Cer: 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
67 |
|
68 |
## Training procedure
|
69 |
|
@@ -86,18 +53,19 @@ The following hyperparameters were used during training:
|
|
86 |
|
87 |
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|
88 |
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
|
89 |
-
|
|
90 |
-
|
|
91 |
-
| 0.
|
92 |
-
| 0.
|
93 |
-
| 0.
|
94 |
-
| 0.
|
95 |
-
| 0.
|
|
|
96 |
|
97 |
|
98 |
### Framework versions
|
99 |
|
100 |
-
- Transformers 4.
|
101 |
-
- Pytorch 1.10.
|
102 |
-
- Datasets 1.
|
103 |
-
- Tokenizers 0.
|
|
|
1 |
---
|
2 |
+
license: mit
|
|
|
|
|
|
|
3 |
tags:
|
4 |
+
- generated_from_trainer
|
|
|
5 |
datasets:
|
6 |
+
- common_voice
|
|
|
|
|
|
|
7 |
model-index:
|
8 |
- name: wav2vec2-large-xlsr-53-punjabi
|
9 |
+
results: []
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
10 |
---
|
11 |
|
12 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
14 |
|
15 |
# wav2vec2-large-xlsr-53-punjabi
|
16 |
|
17 |
+
This model is a fine-tuned version of [Harveenchadha/vakyansh-wav2vec2-punjabi-pam-10](https://huggingface.co/Harveenchadha/vakyansh-wav2vec2-punjabi-pam-10) on the common_voice dataset.
|
18 |
It achieves the following results on the evaluation set:
|
19 |
+
- Loss: 1.2101
|
20 |
+
- Wer: 0.4939
|
21 |
+
- Cer: 0.2238
|
22 |
+
|
23 |
+
## Model description
|
24 |
+
|
25 |
+
More information needed
|
26 |
+
|
27 |
+
## Intended uses & limitations
|
28 |
+
|
29 |
+
More information needed
|
30 |
+
|
31 |
+
## Training and evaluation data
|
32 |
+
|
33 |
+
More information needed
|
34 |
|
35 |
## Training procedure
|
36 |
|
|
|
53 |
|
54 |
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|
55 |
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
|
56 |
+
| 11.0563 | 3.7 | 100 | 1.9492 | 0.7123 | 0.3872 |
|
57 |
+
| 1.6715 | 7.41 | 200 | 1.3142 | 0.6433 | 0.3086 |
|
58 |
+
| 0.9117 | 11.11 | 300 | 1.2733 | 0.5657 | 0.2627 |
|
59 |
+
| 0.666 | 14.81 | 400 | 1.2730 | 0.5598 | 0.2534 |
|
60 |
+
| 0.4225 | 18.52 | 500 | 1.2548 | 0.5300 | 0.2399 |
|
61 |
+
| 0.3209 | 22.22 | 600 | 1.2166 | 0.5229 | 0.2372 |
|
62 |
+
| 0.2678 | 25.93 | 700 | 1.1795 | 0.5041 | 0.2276 |
|
63 |
+
| 0.2088 | 29.63 | 800 | 1.2101 | 0.4939 | 0.2238 |
|
64 |
|
65 |
|
66 |
### Framework versions
|
67 |
|
68 |
+
- Transformers 4.17.0.dev0
|
69 |
+
- Pytorch 1.10.2+cu102
|
70 |
+
- Datasets 1.18.2.dev0
|
71 |
+
- Tokenizers 0.11.0
|