Common voice validated partition (#3669)

* 'validated' split added to CommonVoice

* CommonVoice dataset_infos.json updated

* add dummy validated data

Co-authored-by: Igor Shalyminov <shalymin@amazon.de>
Co-authored-by: Quentin Lhoest <lhoest.q@gmail.com>

Commit from https://github.com/huggingface/datasets/commit/b06bf4ec036b778f1064581ec39a105f855b0af8

common_voice.py

@@ -694,6 +694,14 @@ class CommonVoice(datasets.GeneratorBasedBuilder):
                     "path_to_clips": path_to_clips,
                 },
             ),
+            datasets.SplitGenerator(
+                name="validated",
+                gen_kwargs={
+                    "files": dl_manager.iter_archive(archive),
+                    "filepath": "/".join([path_to_data, "validated.tsv"]),
+                    "path_to_clips": path_to_clips,
+                },
+            ),
             datasets.SplitGenerator(
                 name="invalidated",
                 gen_kwargs={

dummy/ab/6.1.0/dummy_data.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cea291508bf2b3691be7437ce49570b0bd1da9b1540196f7d9027f01bdd29f64
-size 11523
+oid sha256:85e49403f105dd492de707786bfd46d627f7fbf7b3bcb39d06ccf1256272d4a6
+size 12601