Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,43 @@
|
|
| 1 |
-
---
|
| 2 |
-
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- ca
|
| 4 |
+
- cy
|
| 5 |
+
- da
|
| 6 |
+
- de
|
| 7 |
+
- en
|
| 8 |
+
- es
|
| 9 |
+
- et
|
| 10 |
+
- eu
|
| 11 |
+
- fa
|
| 12 |
+
- fr
|
| 13 |
+
- ga
|
| 14 |
+
- hr
|
| 15 |
+
- hu
|
| 16 |
+
- id
|
| 17 |
+
- is
|
| 18 |
+
- it
|
| 19 |
+
- ja
|
| 20 |
+
- ko
|
| 21 |
+
- nl
|
| 22 |
+
- no
|
| 23 |
+
- pl
|
| 24 |
+
- pt
|
| 25 |
+
- qu
|
| 26 |
+
- ro
|
| 27 |
+
- sr
|
| 28 |
+
- sv
|
| 29 |
+
- tr
|
| 30 |
+
- zh
|
| 31 |
+
- yue
|
| 32 |
+
license: cc-by-4.0
|
| 33 |
+
---
|
| 34 |
+
|
| 35 |
+
# IPA-CHILDES split
|
| 36 |
+
|
| 37 |
+
This dataset is a postprocessed version of the [IPA-CHILDES](https://huggingface.co/datasets/phonemetransformers/IPA-CHILDES) dataset. In particular,
|
| 38 |
+
the following changes have been implemented:
|
| 39 |
+
- column `lang` was added to make the language code (ISO 639-1/2) accessible during training/inference time
|
| 40 |
+
- word boundary marker was changed from `WORD_BOUNDARY` to the character `|`
|
| 41 |
+
- a train/test split is now provided
|
| 42 |
+
|
| 43 |
+
The distribution license is inherited from the original dataset and so available to free copy and use under the CC-BY-4.0 license.
|