Datasets:
Update README.md
Browse files
README.md
CHANGED
@@ -70,8 +70,8 @@ dataset_info:
|
|
70 |
# Composite dataset for Basque made from public available data
|
71 |
This dataset is composed of the following public available data:
|
72 |
|
73 |
-
|
74 |
-
|
75 |
- **mozilla-foundation/common_voice_18_0/eu**: "validated" split removing "test_cv" and "dev_cv" split's sentences. (validated split contains official train + dev + test splits and more unique data)
|
76 |
- **gttsehu/basque_parliament_1/eu**: "train_clean" split removing some of the sentences that are repeated in "test_parl" and "dev_parl" splits (not the same recording but same text).
|
77 |
- **openslr**: a train split made from the SLR76 (Basque recordings) subset, this split has been cleaned from acronyms, numbers and sentences that are repeated in the following "test_oslr" and "dev_oslr" splits.
|
@@ -82,8 +82,8 @@ This dataset is composed of the following public available data:
|
|
82 |
| - | openslr | 6.28 h | 3229 |
|
83 |
| train | **Total** | **675.98 h** | **387426** |
|
84 |
|
85 |
-
|
86 |
-
|
87 |
- **mozilla-foundation/common_voice_18_0/eu**: official "test" split.
|
88 |
- **gttsehu/basque_parliament_1/eu**: official "test" split.
|
89 |
- **openslr**: a test split made from the SLR76 (Basque recordings) subset, this split has been cleaned from acronyms, numbers and repeated sentences so all of them are only once in the split.
|
@@ -94,7 +94,7 @@ This dataset is composed of the following public available data:
|
|
94 |
| test_oslr | openslr | 1.03 h | 526 |
|
95 |
| | **Total** | **28.08 h** | **16359** |
|
96 |
|
97 |
-
|
98 |
Theres a dev split composed by 3 dev subsplits that are also independently accesible. It is recommended to use the combined "dev" split for development tasks since it is accurately balanced in number of hours.
|
99 |
- **mozilla-foundation/common_voice_18_0/eu**: a small dev split made from the official "dev" split.
|
100 |
- **gttsehu/basque_parliament_1/eu**: a small dev split made from the official "dev" split.
|
|
|
70 |
# Composite dataset for Basque made from public available data
|
71 |
This dataset is composed of the following public available data:
|
72 |
|
73 |
+
## Train split:
|
74 |
+
The train split is composed of the following datasets combined:
|
75 |
- **mozilla-foundation/common_voice_18_0/eu**: "validated" split removing "test_cv" and "dev_cv" split's sentences. (validated split contains official train + dev + test splits and more unique data)
|
76 |
- **gttsehu/basque_parliament_1/eu**: "train_clean" split removing some of the sentences that are repeated in "test_parl" and "dev_parl" splits (not the same recording but same text).
|
77 |
- **openslr**: a train split made from the SLR76 (Basque recordings) subset, this split has been cleaned from acronyms, numbers and sentences that are repeated in the following "test_oslr" and "dev_oslr" splits.
|
|
|
82 |
| - | openslr | 6.28 h | 3229 |
|
83 |
| train | **Total** | **675.98 h** | **387426** |
|
84 |
|
85 |
+
## Test splits:
|
86 |
+
Those test splits are separated, and it is recommended to not evaluate them together in a single split:
|
87 |
- **mozilla-foundation/common_voice_18_0/eu**: official "test" split.
|
88 |
- **gttsehu/basque_parliament_1/eu**: official "test" split.
|
89 |
- **openslr**: a test split made from the SLR76 (Basque recordings) subset, this split has been cleaned from acronyms, numbers and repeated sentences so all of them are only once in the split.
|
|
|
94 |
| test_oslr | openslr | 1.03 h | 526 |
|
95 |
| | **Total** | **28.08 h** | **16359** |
|
96 |
|
97 |
+
## Dev splits:
|
98 |
Theres a dev split composed by 3 dev subsplits that are also independently accesible. It is recommended to use the combined "dev" split for development tasks since it is accurately balanced in number of hours.
|
99 |
- **mozilla-foundation/common_voice_18_0/eu**: a small dev split made from the official "dev" split.
|
100 |
- **gttsehu/basque_parliament_1/eu**: a small dev split made from the official "dev" split.
|