pierreguillou
commited on
Commit
•
a33134e
1
Parent(s):
7f2ed75
Update README.md
Browse files
README.md
CHANGED
@@ -21,11 +21,11 @@ model-index:
|
|
21 |
args: fr
|
22 |
metrics:
|
23 |
- name: Wer
|
24 |
-
type: wer
|
25 |
-
value: 15.89689189275029
|
26 |
-
- name: Wer norm
|
27 |
type: wer
|
28 |
value: 11.1406
|
|
|
|
|
|
|
29 |
---
|
30 |
|
31 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -36,8 +36,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
36 |
This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the common_voice_11_0 dataset.
|
37 |
It achieves the following results on the evaluation set:
|
38 |
- Loss: 0.2664
|
39 |
-
- Wer: 15.8969
|
40 |
-
- Wer
|
41 |
|
42 |
## New SOTA
|
43 |
|
|
|
21 |
args: fr
|
22 |
metrics:
|
23 |
- name: Wer
|
|
|
|
|
|
|
24 |
type: wer
|
25 |
value: 11.1406
|
26 |
+
- name: Wer (without normalization)
|
27 |
+
type: wer_without_norm
|
28 |
+
value: 15.89689189275029
|
29 |
---
|
30 |
|
31 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
36 |
This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the common_voice_11_0 dataset.
|
37 |
It achieves the following results on the evaluation set:
|
38 |
- Loss: 0.2664
|
39 |
+
- Wer (without normalization): 15.8969
|
40 |
+
- Wer (with normalization): **11.1406**
|
41 |
|
42 |
## New SOTA
|
43 |
|