Update the new v2 split
Browse filesUpdating the split.
Created training split
and reduced the dev and eval sessions.
README.md
CHANGED
@@ -42,12 +42,15 @@ Example) `session_ge1nse2c.seglst.json`
|
|
42 |
|
43 |
The dataset is divided into training and test splits:
|
44 |
|
45 |
-
-
|
46 |
-
- 2 to 4 speakers in each session
|
47 |
-
- Approximately 10 ~ 40 mins of recordings
|
48 |
-
- Evaluation Data: 104 entries
|
49 |
- 2 to 4 speakers in each session
|
50 |
- Approximately 10 ~ 40 mins of recordings
|
|
|
|
|
|
|
|
|
|
|
|
|
51 |
|
52 |
### Keys (items)
|
53 |
|
@@ -62,15 +65,18 @@ The dataset is divided into training and test splits:
|
|
62 |
`err_source_text`: This is the erroneous ASR-Diarization results to be fixed. Has dev, eval folders
|
63 |
`ref_annotated_text`: This is the human annotated ground-truth for evaluation. Only dev split is included.
|
64 |
|
|
|
|
|
|
|
65 |
- **Development Sources**:
|
66 |
-
- `dev`:
|
67 |
|
68 |
- **Evaluation Sources**:
|
69 |
-
- `eval`:
|
70 |
|
71 |
## Access
|
72 |
|
73 |
-
The dataset can be accessed and downloaded through the HuggingFace Datasets library.
|
74 |
|
75 |
## Evaluation
|
76 |
|
|
|
42 |
|
43 |
The dataset is divided into training and test splits:
|
44 |
|
45 |
+
- Training Data: 222 entries
|
|
|
|
|
|
|
46 |
- 2 to 4 speakers in each session
|
47 |
- Approximately 10 ~ 40 mins of recordings
|
48 |
+
- Development Data: 13 entries
|
49 |
+
- 2 speakers in each session
|
50 |
+
- Approximately 10 mins of recordings
|
51 |
+
- Evaluation Data: 11 entries
|
52 |
+
- 2 speakers in each session
|
53 |
+
- Approximately 10 mins of recordings
|
54 |
|
55 |
### Keys (items)
|
56 |
|
|
|
65 |
`err_source_text`: This is the erroneous ASR-Diarization results to be fixed. Has dev, eval folders
|
66 |
`ref_annotated_text`: This is the human annotated ground-truth for evaluation. Only dev split is included.
|
67 |
|
68 |
+
- **Training Sources**:
|
69 |
+
- `dev`: 222 sessions
|
70 |
+
|
71 |
- **Development Sources**:
|
72 |
+
- `dev`: 13 sessions
|
73 |
|
74 |
- **Evaluation Sources**:
|
75 |
+
- `eval`: 11 Sessions
|
76 |
|
77 |
## Access
|
78 |
|
79 |
+
The dataset can be accessed and downloaded through the HuggingFace Datasets library (i.e., This Repository).
|
80 |
|
81 |
## Evaluation
|
82 |
|