Hugging Face's logo

Spaces:
diarizers-community
/
README
Running

App Files Community
kamilakesbi commited on
Commit
f20db24
·
verified ·
1 Parent(s): c6cbee4

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +14 -15
README.md CHANGED
@@ -18,21 +18,20 @@ The available datasets are the CallHome (Japanese, Chinese, German, Spanish, Eng
18
  Each model has been fine-tuned on a specific Callhome language subset. They achieve better performances on multilingual data compared to pyannote's pre-trained [segmentation-3.0](https://huggingface.co/pyannote/segmentation-3.0) model:
19
 
20
 
21
- | Callhome test dataset subset | Model | DER | False alarm | Missed detection| Confusion |
22
- | ------------------------| ------------- | ------------- | ------------- | --------------- | ------------- |
23
- | Japanese | Pretrained | 25.44 | **2.30** | 17.45 | 5.69 |
24
- | | Fine-tuned | **18.23** | 6.31 | **6.91** | **5.01** |
25
- | Spanish | Pretrained | 33.44 | **2.59** | 25.19 | **5.66** |
26
- | | Fine-tuned | **25.72** | 6.87 | **12.73** | 6.12 |
27
- | English | Pretrained | 22.16 | **6.29** | 10.97 | 4.90 |
28
- | | Fine-tuned | **18.40** | 7.10 | **6.98** | **4.32** |
29
- | German | Pretrained | 21.90 | **3.10** | 14.25 | 4.55 |
30
- | | Fine-tuned | **16.75** | 5.00 | **7.75** | **4.00** |
31
- | Chinese | Pretrained | 19.73 | **4.81** | 9.82 | 5.11 |
32
- | | Fine-tuned | **15.95** | 5.04 | **7.24** | **3.68** |
33
-
34
-
35
- Note: Results have been obtained using the [test script](https://github.com/kamilakesbi/diarizers/blob/main/test_segmentation.py) from diarizers.
36
 
37
  Together with diarizers-community, we release:
38
 
 
18
  Each model has been fine-tuned on a specific Callhome language subset. They achieve better performances on multilingual data compared to pyannote's pre-trained [segmentation-3.0](https://huggingface.co/pyannote/segmentation-3.0) model:
19
 
20
 
21
+ | [Callhome](https://huggingface.co/datasets/diarizers-community/callhome) test dataset subset| Model | DER | False alarm | Missed detection| Confusion |
22
+ | ------------------------| ------------- | ------------- | ------------- | --------------- | ------------- |
23
+ | Japanese | [Pretrained](https://huggingface.co/pyannote/segmentation-3.0) | 25.44 | **2.30** | 17.45 | 5.69 |
24
+ | | [Fine-tuned](https://huggingface.co/diarizers-community/speaker-segmentation-fine-tuned-callhome-jpn) | **18.23** | 6.31 | **6.91** | **5.01** |
25
+ | Spanish | [Pretrained](https://huggingface.co/pyannote/segmentation-3.0) | 33.44 | **2.59** | 25.19 | **5.66** |
26
+ | | [Fine-tuned](https://huggingface.co/diarizers-community/speaker-segmentation-fine-tuned-callhome-spa) | **25.72** | 6.87 | **12.73** | 6.12 |
27
+ | English | [Pretrained](https://huggingface.co/pyannote/segmentation-3.0) | 22.16 | **6.29** | 10.97 | 4.90 |
28
+ | | [Fine-tuned](https://huggingface.co/diarizers-community/speaker-segmentation-fine-tuned-callhome-eng) | **18.40** | 7.10 | **6.98** | **4.32** |
29
+ | German | [Pretrained](https://huggingface.co/pyannote/segmentation-3.0) | 21.90 | **3.10** | 14.25 | 4.55 |
30
+ | | [Fine-tuned](https://huggingface.co/diarizers-community/speaker-segmentation-fine-tuned-callhome-deu) | **16.75** | 5.00 | **7.75** | **4.00** |
31
+ | Chinese | [Pretrained](https://huggingface.co/pyannote/segmentation-3.0) | 19.73 | **4.81** | 9.82 | 5.11 |
32
+ | | [Fine-tuned](https://huggingface.co/diarizers-community/speaker-segmentation-fine-tuned-callhome-zho) | **15.95** | 5.04 | **7.24** | **3.68** |
33
+
34
+ Results are in %. They have been obtained using the [test script](https://github.com/kamilakesbi/diarizers/blob/main/test_segmentation.py) from diarizers.
 
35
 
36
  Together with diarizers-community, we release:
37