update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,123 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
model-index:
|
6 |
+
- name: kids_phoneme_sm_model
|
7 |
+
results: []
|
8 |
+
---
|
9 |
+
|
10 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
11 |
+
should probably proofread and complete it, then remove this comment. -->
|
12 |
+
|
13 |
+
# kids_phoneme_sm_model
|
14 |
+
|
15 |
+
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the None dataset.
|
16 |
+
It achieves the following results on the evaluation set:
|
17 |
+
- Loss: 1.4558
|
18 |
+
- Cer: 0.4079
|
19 |
+
|
20 |
+
## Model description
|
21 |
+
|
22 |
+
More information needed
|
23 |
+
|
24 |
+
## Intended uses & limitations
|
25 |
+
|
26 |
+
More information needed
|
27 |
+
|
28 |
+
## Training and evaluation data
|
29 |
+
|
30 |
+
More information needed
|
31 |
+
|
32 |
+
## Training procedure
|
33 |
+
|
34 |
+
### Training hyperparameters
|
35 |
+
|
36 |
+
The following hyperparameters were used during training:
|
37 |
+
- learning_rate: 0.0001
|
38 |
+
- train_batch_size: 2
|
39 |
+
- eval_batch_size: 8
|
40 |
+
- seed: 42
|
41 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
42 |
+
- lr_scheduler_type: linear
|
43 |
+
- num_epochs: 50
|
44 |
+
|
45 |
+
### Training results
|
46 |
+
|
47 |
+
| Training Loss | Epoch | Step | Validation Loss | Cer |
|
48 |
+
|:-------------:|:-----:|:-----:|:---------------:|:------:|
|
49 |
+
| 3.0642 | 0.74 | 500 | 4.4995 | 1.0 |
|
50 |
+
| 2.8486 | 1.48 | 1000 | 3.8639 | 1.0 |
|
51 |
+
| 2.7909 | 2.22 | 1500 | 3.4712 | 1.0 |
|
52 |
+
| 1.5475 | 2.96 | 2000 | 1.0263 | 0.6825 |
|
53 |
+
| 0.7353 | 3.7 | 2500 | 0.8291 | 0.5760 |
|
54 |
+
| 0.6036 | 4.44 | 3000 | 0.7387 | 0.5327 |
|
55 |
+
| 0.5553 | 5.19 | 3500 | 0.7382 | 0.5023 |
|
56 |
+
| 0.4271 | 5.93 | 4000 | 0.7244 | 0.4991 |
|
57 |
+
| 0.43 | 6.67 | 4500 | 0.7152 | 0.4805 |
|
58 |
+
| 0.3925 | 7.41 | 5000 | 0.7210 | 0.4587 |
|
59 |
+
| 0.3719 | 8.15 | 5500 | 0.7888 | 0.4491 |
|
60 |
+
| 0.3451 | 8.89 | 6000 | 0.7599 | 0.4433 |
|
61 |
+
| 0.319 | 9.63 | 6500 | 0.7642 | 0.4508 |
|
62 |
+
| 0.2638 | 10.37 | 7000 | 0.8490 | 0.4426 |
|
63 |
+
| 0.3084 | 11.11 | 7500 | 0.9387 | 0.4315 |
|
64 |
+
| 0.2553 | 11.85 | 8000 | 0.8477 | 0.4287 |
|
65 |
+
| 0.2537 | 12.59 | 8500 | 0.8261 | 0.4301 |
|
66 |
+
| 0.2058 | 13.33 | 9000 | 1.1093 | 0.4247 |
|
67 |
+
| 0.2283 | 14.07 | 9500 | 0.7638 | 0.4230 |
|
68 |
+
| 0.2043 | 14.81 | 10000 | 1.0104 | 0.4219 |
|
69 |
+
| 0.1918 | 15.56 | 10500 | 0.9618 | 0.4194 |
|
70 |
+
| 0.1764 | 16.3 | 11000 | 0.9460 | 0.4226 |
|
71 |
+
| 0.1677 | 17.04 | 11500 | 0.9750 | 0.4233 |
|
72 |
+
| 0.1751 | 17.78 | 12000 | 0.9600 | 0.4240 |
|
73 |
+
| 0.1465 | 18.52 | 12500 | 1.1328 | 0.4172 |
|
74 |
+
| 0.1239 | 19.26 | 13000 | 1.0746 | 0.4176 |
|
75 |
+
| 0.1495 | 20.0 | 13500 | 1.2143 | 0.4194 |
|
76 |
+
| 0.1444 | 20.74 | 14000 | 1.1595 | 0.4219 |
|
77 |
+
| 0.134 | 21.48 | 14500 | 1.1601 | 0.4201 |
|
78 |
+
| 0.1343 | 22.22 | 15000 | 1.1730 | 0.4233 |
|
79 |
+
| 0.1051 | 22.96 | 15500 | 1.1257 | 0.4172 |
|
80 |
+
| 0.1067 | 23.7 | 16000 | 1.1206 | 0.4190 |
|
81 |
+
| 0.0959 | 24.44 | 16500 | 1.1539 | 0.4133 |
|
82 |
+
| 0.1028 | 25.19 | 17000 | 1.2425 | 0.4126 |
|
83 |
+
| 0.1028 | 25.93 | 17500 | 1.2008 | 0.4144 |
|
84 |
+
| 0.1052 | 26.67 | 18000 | 1.1974 | 0.4094 |
|
85 |
+
| 0.0813 | 27.41 | 18500 | 1.0960 | 0.4133 |
|
86 |
+
| 0.0973 | 28.15 | 19000 | 1.1153 | 0.4101 |
|
87 |
+
| 0.0783 | 28.89 | 19500 | 1.1596 | 0.4126 |
|
88 |
+
| 0.0704 | 29.63 | 20000 | 1.1881 | 0.4087 |
|
89 |
+
| 0.068 | 30.37 | 20500 | 1.2289 | 0.4040 |
|
90 |
+
| 0.0664 | 31.11 | 21000 | 1.2289 | 0.4079 |
|
91 |
+
| 0.0747 | 31.85 | 21500 | 1.2642 | 0.4122 |
|
92 |
+
| 0.0663 | 32.59 | 22000 | 1.3062 | 0.4101 |
|
93 |
+
| 0.0668 | 33.33 | 22500 | 1.3486 | 0.4101 |
|
94 |
+
| 0.0592 | 34.07 | 23000 | 1.3346 | 0.4040 |
|
95 |
+
| 0.0513 | 34.81 | 23500 | 1.2958 | 0.4097 |
|
96 |
+
| 0.0511 | 35.56 | 24000 | 1.3798 | 0.4108 |
|
97 |
+
| 0.0557 | 36.3 | 24500 | 1.3521 | 0.4065 |
|
98 |
+
| 0.049 | 37.04 | 25000 | 1.4192 | 0.4094 |
|
99 |
+
| 0.0465 | 37.78 | 25500 | 1.4308 | 0.4108 |
|
100 |
+
| 0.0474 | 38.52 | 26000 | 1.4004 | 0.4058 |
|
101 |
+
| 0.0428 | 39.26 | 26500 | 1.3988 | 0.4054 |
|
102 |
+
| 0.0509 | 40.0 | 27000 | 1.4218 | 0.4069 |
|
103 |
+
| 0.0386 | 40.74 | 27500 | 1.3819 | 0.4104 |
|
104 |
+
| 0.0426 | 41.48 | 28000 | 1.4681 | 0.4090 |
|
105 |
+
| 0.0408 | 42.22 | 28500 | 1.4543 | 0.4104 |
|
106 |
+
| 0.0405 | 42.96 | 29000 | 1.4999 | 0.4108 |
|
107 |
+
| 0.036 | 43.7 | 29500 | 1.4922 | 0.4072 |
|
108 |
+
| 0.036 | 44.44 | 30000 | 1.4709 | 0.4087 |
|
109 |
+
| 0.04 | 45.19 | 30500 | 1.4858 | 0.4094 |
|
110 |
+
| 0.0343 | 45.93 | 31000 | 1.4606 | 0.4087 |
|
111 |
+
| 0.0288 | 46.67 | 31500 | 1.4599 | 0.4044 |
|
112 |
+
| 0.0454 | 47.41 | 32000 | 1.4288 | 0.4087 |
|
113 |
+
| 0.0322 | 48.15 | 32500 | 1.4589 | 0.4083 |
|
114 |
+
| 0.0327 | 48.89 | 33000 | 1.4502 | 0.4094 |
|
115 |
+
| 0.0272 | 49.63 | 33500 | 1.4558 | 0.4079 |
|
116 |
+
|
117 |
+
|
118 |
+
### Framework versions
|
119 |
+
|
120 |
+
- Transformers 4.30.1
|
121 |
+
- Pytorch 2.0.0
|
122 |
+
- Datasets 2.12.0
|
123 |
+
- Tokenizers 0.13.3
|