DrishtiSharma
commited on
Commit
•
07fb825
1
Parent(s):
64f56f5
update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- ab
|
4 |
+
license: apache-2.0
|
5 |
+
tags:
|
6 |
+
- automatic-speech-recognition
|
7 |
+
- mozilla-foundation/common_voice_7_0
|
8 |
+
- generated_from_trainer
|
9 |
+
datasets:
|
10 |
+
- common_voice
|
11 |
+
model-index:
|
12 |
+
- name: ''
|
13 |
+
results: []
|
14 |
+
---
|
15 |
+
|
16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
17 |
+
should probably proofread and complete it, then remove this comment. -->
|
18 |
+
|
19 |
+
#
|
20 |
+
|
21 |
+
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the MOZILLA-FOUNDATION/COMMON_VOICE_7_0 - AB dataset.
|
22 |
+
It achieves the following results on the evaluation set:
|
23 |
+
- Loss: 0.5885
|
24 |
+
- Wer: 0.5899
|
25 |
+
|
26 |
+
## Model description
|
27 |
+
|
28 |
+
More information needed
|
29 |
+
|
30 |
+
## Intended uses & limitations
|
31 |
+
|
32 |
+
More information needed
|
33 |
+
|
34 |
+
## Training and evaluation data
|
35 |
+
|
36 |
+
More information needed
|
37 |
+
|
38 |
+
## Training procedure
|
39 |
+
|
40 |
+
### Training hyperparameters
|
41 |
+
|
42 |
+
The following hyperparameters were used during training:
|
43 |
+
- learning_rate: 7.5e-05
|
44 |
+
- train_batch_size: 16
|
45 |
+
- eval_batch_size: 8
|
46 |
+
- seed: 42
|
47 |
+
- gradient_accumulation_steps: 2
|
48 |
+
- total_train_batch_size: 32
|
49 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
50 |
+
- lr_scheduler_type: linear
|
51 |
+
- lr_scheduler_warmup_steps: 750
|
52 |
+
- num_epochs: 100.0
|
53 |
+
- mixed_precision_training: Native AMP
|
54 |
+
|
55 |
+
### Training results
|
56 |
+
|
57 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
58 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
59 |
+
| 7.5798 | 13.64 | 300 | 3.4349 | 1.0 |
|
60 |
+
| 3.1252 | 27.27 | 600 | 3.0706 | 1.0 |
|
61 |
+
| 2.2546 | 40.91 | 900 | 0.8427 | 0.7763 |
|
62 |
+
| 0.7564 | 54.55 | 1200 | 0.6129 | 0.6376 |
|
63 |
+
| 0.5239 | 68.18 | 1500 | 0.5769 | 0.6037 |
|
64 |
+
| 0.438 | 81.82 | 1800 | 0.5938 | 0.5915 |
|
65 |
+
| 0.3945 | 95.45 | 2100 | 0.5869 | 0.5861 |
|
66 |
+
|
67 |
+
|
68 |
+
### Framework versions
|
69 |
+
|
70 |
+
- Transformers 4.16.0.dev0
|
71 |
+
- Pytorch 1.10.1+cu102
|
72 |
+
- Datasets 1.17.1.dev0
|
73 |
+
- Tokenizers 0.11.0
|