update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,104 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: cc-by-nc-4.0
|
3 |
+
base_model: facebook/mms-1b-fl102
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
datasets:
|
7 |
+
- common_voice_6_1
|
8 |
+
metrics:
|
9 |
+
- wer
|
10 |
+
model-index:
|
11 |
+
- name: wav2vec2-large-mms-1b-hindi-colab
|
12 |
+
results:
|
13 |
+
- task:
|
14 |
+
name: Automatic Speech Recognition
|
15 |
+
type: automatic-speech-recognition
|
16 |
+
dataset:
|
17 |
+
name: common_voice_6_1
|
18 |
+
type: common_voice_6_1
|
19 |
+
config: hi
|
20 |
+
split: test
|
21 |
+
args: hi
|
22 |
+
metrics:
|
23 |
+
- name: Wer
|
24 |
+
type: wer
|
25 |
+
value: 0.32018561484918795
|
26 |
+
---
|
27 |
+
|
28 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
29 |
+
should probably proofread and complete it, then remove this comment. -->
|
30 |
+
|
31 |
+
# wav2vec2-large-mms-1b-hindi-colab
|
32 |
+
|
33 |
+
This model is a fine-tuned version of [facebook/mms-1b-fl102](https://huggingface.co/facebook/mms-1b-fl102) on the common_voice_6_1 dataset.
|
34 |
+
It achieves the following results on the evaluation set:
|
35 |
+
- Loss: 0.3535
|
36 |
+
- Wer: 0.3202
|
37 |
+
|
38 |
+
## Model description
|
39 |
+
|
40 |
+
More information needed
|
41 |
+
|
42 |
+
## Intended uses & limitations
|
43 |
+
|
44 |
+
More information needed
|
45 |
+
|
46 |
+
## Training and evaluation data
|
47 |
+
|
48 |
+
More information needed
|
49 |
+
|
50 |
+
## Training procedure
|
51 |
+
|
52 |
+
### Training hyperparameters
|
53 |
+
|
54 |
+
The following hyperparameters were used during training:
|
55 |
+
- learning_rate: 0.001
|
56 |
+
- train_batch_size: 4
|
57 |
+
- eval_batch_size: 8
|
58 |
+
- seed: 42
|
59 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
60 |
+
- lr_scheduler_type: linear
|
61 |
+
- lr_scheduler_warmup_steps: 10
|
62 |
+
- num_epochs: 4
|
63 |
+
|
64 |
+
### Training results
|
65 |
+
|
66 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
67 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
68 |
+
| 16.7585 | 0.14 | 10 | 10.2106 | 2.0951 |
|
69 |
+
| 6.9602 | 0.27 | 20 | 3.7700 | 1.0046 |
|
70 |
+
| 2.4653 | 0.41 | 30 | 1.3321 | 0.6763 |
|
71 |
+
| 1.0919 | 0.55 | 40 | 0.6594 | 0.4664 |
|
72 |
+
| 0.7645 | 0.68 | 50 | 0.4930 | 0.3910 |
|
73 |
+
| 0.8434 | 0.82 | 60 | 0.4819 | 0.3898 |
|
74 |
+
| 0.5118 | 0.96 | 70 | 0.4492 | 0.3817 |
|
75 |
+
| 0.6097 | 1.1 | 80 | 0.4299 | 0.4327 |
|
76 |
+
| 0.4698 | 1.23 | 90 | 0.4308 | 0.3643 |
|
77 |
+
| 0.5402 | 1.37 | 100 | 0.4042 | 0.4107 |
|
78 |
+
| 0.5622 | 1.51 | 110 | 0.4156 | 0.3701 |
|
79 |
+
| 0.4084 | 1.64 | 120 | 0.4138 | 0.3701 |
|
80 |
+
| 0.4888 | 1.78 | 130 | 0.3917 | 0.3434 |
|
81 |
+
| 0.4253 | 1.92 | 140 | 0.3852 | 0.3457 |
|
82 |
+
| 0.5004 | 2.05 | 150 | 0.3843 | 0.3364 |
|
83 |
+
| 0.3791 | 2.19 | 160 | 0.3841 | 0.3469 |
|
84 |
+
| 0.3302 | 2.33 | 170 | 0.3764 | 0.3271 |
|
85 |
+
| 0.4047 | 2.47 | 180 | 0.3689 | 0.3364 |
|
86 |
+
| 0.2951 | 2.6 | 190 | 0.3657 | 0.3329 |
|
87 |
+
| 0.3545 | 2.74 | 200 | 0.3582 | 0.3306 |
|
88 |
+
| 0.3736 | 2.88 | 210 | 0.3585 | 0.3248 |
|
89 |
+
| 0.388 | 3.01 | 220 | 0.3602 | 0.3237 |
|
90 |
+
| 0.2997 | 3.15 | 230 | 0.3624 | 0.3167 |
|
91 |
+
| 0.3704 | 3.29 | 240 | 0.3625 | 0.3190 |
|
92 |
+
| 0.2095 | 3.42 | 250 | 0.3571 | 0.3248 |
|
93 |
+
| 0.3564 | 3.56 | 260 | 0.3570 | 0.3202 |
|
94 |
+
| 0.2119 | 3.7 | 270 | 0.3550 | 0.3225 |
|
95 |
+
| 0.3697 | 3.84 | 280 | 0.3542 | 0.3190 |
|
96 |
+
| 0.3551 | 3.97 | 290 | 0.3535 | 0.3202 |
|
97 |
+
|
98 |
+
|
99 |
+
### Framework versions
|
100 |
+
|
101 |
+
- Transformers 4.31.0.dev0
|
102 |
+
- Pytorch 2.0.1+cu118
|
103 |
+
- Datasets 2.13.1
|
104 |
+
- Tokenizers 0.13.3
|