update model card README.md
Browse files
README.md
CHANGED
@@ -16,12 +16,12 @@ model-index:
|
|
16 |
name: common_voice_8_0
|
17 |
type: common_voice_8_0
|
18 |
config: sw
|
19 |
-
split: test[:
|
20 |
args: sw
|
21 |
metrics:
|
22 |
- name: Wer
|
23 |
type: wer
|
24 |
-
value:
|
25 |
---
|
26 |
|
27 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -31,8 +31,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
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_8_0 dataset.
|
33 |
It achieves the following results on the evaluation set:
|
34 |
-
- Loss:
|
35 |
-
- Wer:
|
36 |
|
37 |
## Model description
|
38 |
|
@@ -60,16 +60,24 @@ The following hyperparameters were used during training:
|
|
60 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
61 |
- lr_scheduler_type: linear
|
62 |
- lr_scheduler_warmup_steps: 50
|
63 |
-
- num_epochs:
|
64 |
|
65 |
### Training results
|
66 |
|
67 |
-
| Training Loss | Epoch | Step | Validation Loss | Wer
|
68 |
-
|
69 |
-
| 6.
|
70 |
-
| 2.
|
71 |
-
| 2.
|
72 |
-
| 2.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
73 |
|
74 |
|
75 |
### Framework versions
|
|
|
16 |
name: common_voice_8_0
|
17 |
type: common_voice_8_0
|
18 |
config: sw
|
19 |
+
split: test[:400]
|
20 |
args: sw
|
21 |
metrics:
|
22 |
- name: Wer
|
23 |
type: wer
|
24 |
+
value: 0.97
|
25 |
---
|
26 |
|
27 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
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_8_0 dataset.
|
33 |
It achieves the following results on the evaluation set:
|
34 |
+
- Loss: 0.4900
|
35 |
+
- Wer: 0.97
|
36 |
|
37 |
## Model description
|
38 |
|
|
|
60 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
61 |
- lr_scheduler_type: linear
|
62 |
- lr_scheduler_warmup_steps: 50
|
63 |
+
- num_epochs: 5
|
64 |
|
65 |
### Training results
|
66 |
|
67 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
68 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
69 |
+
| 6.5497 | 0.4 | 50 | 2.9819 | 1.0 |
|
70 |
+
| 2.8809 | 0.8 | 100 | 2.8873 | 1.0 |
|
71 |
+
| 2.8416 | 1.2 | 150 | 2.8427 | 1.0 |
|
72 |
+
| 2.8145 | 1.6 | 200 | 2.8067 | 1.0 |
|
73 |
+
| 2.747 | 2.0 | 250 | 2.7092 | 1.0 |
|
74 |
+
| 2.1095 | 2.4 | 300 | 1.3472 | 1.0 |
|
75 |
+
| 0.9546 | 2.8 | 350 | 0.7708 | 0.9975 |
|
76 |
+
| 0.6104 | 3.2 | 400 | 0.6317 | 0.9825 |
|
77 |
+
| 0.4941 | 3.6 | 450 | 0.5427 | 0.97 |
|
78 |
+
| 0.4345 | 4.0 | 500 | 0.5314 | 0.975 |
|
79 |
+
| 0.3327 | 4.4 | 550 | 0.4927 | 0.9625 |
|
80 |
+
| 0.3099 | 4.8 | 600 | 0.4900 | 0.97 |
|
81 |
|
82 |
|
83 |
### Framework versions
|