Commit
•
77da528
1
Parent(s):
e545385
update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,116 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
datasets:
|
6 |
+
- common_voice_13_0
|
7 |
+
metrics:
|
8 |
+
- wer
|
9 |
+
model-index:
|
10 |
+
- name: b21-wav2vec2-large-xls-r-romansh-colab
|
11 |
+
results:
|
12 |
+
- task:
|
13 |
+
name: Automatic Speech Recognition
|
14 |
+
type: automatic-speech-recognition
|
15 |
+
dataset:
|
16 |
+
name: common_voice_13_0
|
17 |
+
type: common_voice_13_0
|
18 |
+
config: rm-vallader
|
19 |
+
split: test
|
20 |
+
args: rm-vallader
|
21 |
+
metrics:
|
22 |
+
- name: Wer
|
23 |
+
type: wer
|
24 |
+
value: 0.6304145319049836
|
25 |
+
---
|
26 |
+
|
27 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
28 |
+
should probably proofread and complete it, then remove this comment. -->
|
29 |
+
|
30 |
+
# b21-wav2vec2-large-xls-r-romansh-colab
|
31 |
+
|
32 |
+
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the common_voice_13_0 dataset.
|
33 |
+
It achieves the following results on the evaluation set:
|
34 |
+
- Loss: 0.8091
|
35 |
+
- Wer: 0.6304
|
36 |
+
|
37 |
+
## Model description
|
38 |
+
|
39 |
+
More information needed
|
40 |
+
|
41 |
+
## Intended uses & limitations
|
42 |
+
|
43 |
+
More information needed
|
44 |
+
|
45 |
+
## Training and evaluation data
|
46 |
+
|
47 |
+
More information needed
|
48 |
+
|
49 |
+
## Training procedure
|
50 |
+
|
51 |
+
### Training hyperparameters
|
52 |
+
|
53 |
+
The following hyperparameters were used during training:
|
54 |
+
- learning_rate: 0.0004
|
55 |
+
- train_batch_size: 4
|
56 |
+
- eval_batch_size: 8
|
57 |
+
- seed: 42
|
58 |
+
- gradient_accumulation_steps: 2
|
59 |
+
- total_train_batch_size: 8
|
60 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
61 |
+
- lr_scheduler_type: linear
|
62 |
+
- lr_scheduler_warmup_steps: 100
|
63 |
+
- num_epochs: 30
|
64 |
+
- mixed_precision_training: Native AMP
|
65 |
+
|
66 |
+
### Training results
|
67 |
+
|
68 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
69 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
70 |
+
| 6.5829 | 0.76 | 100 | 2.9564 | 1.0 |
|
71 |
+
| 2.9568 | 1.52 | 200 | 3.0768 | 1.0 |
|
72 |
+
| 2.9578 | 2.29 | 300 | 3.0654 | 1.0 |
|
73 |
+
| 2.957 | 3.05 | 400 | 2.9377 | 1.0 |
|
74 |
+
| 2.9419 | 3.81 | 500 | 2.9408 | 1.0 |
|
75 |
+
| 2.9567 | 4.58 | 600 | 2.9395 | 1.0 |
|
76 |
+
| 2.9625 | 5.34 | 700 | 2.9388 | 1.0 |
|
77 |
+
| 2.9395 | 6.11 | 800 | 2.9374 | 1.0 |
|
78 |
+
| 2.9285 | 6.87 | 900 | 2.9240 | 1.0 |
|
79 |
+
| 2.9187 | 7.63 | 1000 | 2.9057 | 1.0 |
|
80 |
+
| 2.9251 | 8.4 | 1100 | 2.8985 | 1.0 |
|
81 |
+
| 2.9033 | 9.16 | 1200 | 2.8942 | 1.0 |
|
82 |
+
| 2.8877 | 9.92 | 1300 | 2.8917 | 1.0 |
|
83 |
+
| 2.8586 | 10.68 | 1400 | 2.7719 | 1.0 |
|
84 |
+
| 2.5777 | 11.45 | 1500 | 2.2424 | 1.0 |
|
85 |
+
| 1.9243 | 12.21 | 1600 | 1.7068 | 0.9772 |
|
86 |
+
| 1.4534 | 12.97 | 1700 | 1.2780 | 0.9585 |
|
87 |
+
| 1.1793 | 13.74 | 1800 | 1.1482 | 0.9360 |
|
88 |
+
| 1.0026 | 14.5 | 1900 | 1.0673 | 0.8852 |
|
89 |
+
| 0.8879 | 15.27 | 2000 | 0.9651 | 0.8433 |
|
90 |
+
| 0.7933 | 16.03 | 2100 | 0.8973 | 0.8216 |
|
91 |
+
| 0.6895 | 16.79 | 2200 | 0.8396 | 0.8034 |
|
92 |
+
| 0.6531 | 17.56 | 2300 | 0.8131 | 0.7713 |
|
93 |
+
| 0.5753 | 18.32 | 2400 | 0.8388 | 0.7531 |
|
94 |
+
| 0.5621 | 19.08 | 2500 | 0.7844 | 0.7632 |
|
95 |
+
| 0.5076 | 19.84 | 2600 | 0.7629 | 0.7485 |
|
96 |
+
| 0.4672 | 20.61 | 2700 | 0.7777 | 0.7497 |
|
97 |
+
| 0.443 | 21.37 | 2800 | 0.8001 | 0.7292 |
|
98 |
+
| 0.4129 | 22.14 | 2900 | 0.7902 | 0.7094 |
|
99 |
+
| 0.3767 | 22.9 | 3000 | 0.7569 | 0.6784 |
|
100 |
+
| 0.357 | 23.66 | 3100 | 0.7726 | 0.6903 |
|
101 |
+
| 0.3378 | 24.43 | 3200 | 0.8016 | 0.6882 |
|
102 |
+
| 0.3199 | 25.19 | 3300 | 0.7854 | 0.6677 |
|
103 |
+
| 0.3144 | 25.95 | 3400 | 0.7792 | 0.6509 |
|
104 |
+
| 0.3025 | 26.71 | 3500 | 0.8157 | 0.6695 |
|
105 |
+
| 0.2919 | 27.48 | 3600 | 0.8215 | 0.6633 |
|
106 |
+
| 0.2762 | 28.24 | 3700 | 0.8167 | 0.6500 |
|
107 |
+
| 0.2679 | 29.01 | 3800 | 0.8144 | 0.6311 |
|
108 |
+
| 0.2671 | 29.77 | 3900 | 0.8091 | 0.6304 |
|
109 |
+
|
110 |
+
|
111 |
+
### Framework versions
|
112 |
+
|
113 |
+
- Transformers 4.26.0
|
114 |
+
- Pytorch 2.0.1+cu118
|
115 |
+
- Datasets 2.14.4
|
116 |
+
- Tokenizers 0.13.3
|