Update README.md
Browse files
README.md
CHANGED
@@ -19,7 +19,18 @@ model-index:
|
|
19 |
name: Speech Recognition
|
20 |
type: automatic-speech-recognition
|
21 |
dataset:
|
22 |
-
name: Common Voice 13.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
23 |
type: common_voice_13_0
|
24 |
args: fa
|
25 |
metrics:
|
@@ -41,8 +52,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
41 |
|
42 |
# wav2vec2-large-xlsr-persian-asr-shemo_me7494
|
43 |
|
44 |
-
This model is a fine-tuned version of [masoudmzb/wav2vec2-xlsr-multilingual-53-fa](https://huggingface.co/masoudmzb/wav2vec2-xlsr-multilingual-53-fa)
|
45 |
-
|
|
|
46 |
- Loss on ShEMO train set: 0.7618
|
47 |
- Loss on ShEMO dev set: 0.6728
|
48 |
- WER on ShEMO train set: 30.47
|
|
|
19 |
name: Speech Recognition
|
20 |
type: automatic-speech-recognition
|
21 |
dataset:
|
22 |
+
name: Common Voice 13.0 fa
|
23 |
+
type: common_voice_13_0
|
24 |
+
args: fa
|
25 |
+
metrics:
|
26 |
+
- name: Test WER
|
27 |
+
type: wer
|
28 |
+
value: 19.21
|
29 |
+
- task:
|
30 |
+
name: Speech Recognition
|
31 |
+
type: automatic-speech-recognition
|
32 |
+
dataset:
|
33 |
+
name: Common Voice 13.0 fa
|
34 |
type: common_voice_13_0
|
35 |
args: fa
|
36 |
metrics:
|
|
|
52 |
|
53 |
# wav2vec2-large-xlsr-persian-asr-shemo_me7494
|
54 |
|
55 |
+
This model is a fine-tuned version of [masoudmzb/wav2vec2-xlsr-multilingual-53-fa](https://huggingface.co/masoudmzb/wav2vec2-xlsr-multilingual-53-fa)
|
56 |
+
on the [ShEMO](https://github.com/pariajm/sharif-emotional-speech-dataset) dataset.
|
57 |
+
It achieves the following results:
|
58 |
- Loss on ShEMO train set: 0.7618
|
59 |
- Loss on ShEMO dev set: 0.6728
|
60 |
- WER on ShEMO train set: 30.47
|