fix wolof dev transcripts

aloresb.py

@@ -109,7 +109,8 @@ class Aloresb(datasets.GeneratorBasedBuilder):
             ),
         ]
         if self.config.name in ["hausa", "wolof"]:
-            train_splits.append(
+            print('IIIIIIIIII')
+            return train_splits + [
                 datasets.SplitGenerator(
                     name=datasets.Split.VALIDATION, gen_kwargs={
                         "transcript_path": transcripts["dev"],
@@ -117,7 +118,15 @@ class Aloresb(datasets.GeneratorBasedBuilder):
                         "local_extracted_archive": local_extracted_archives.get("dev"),
                     }
                 ),
-            )
+                datasets.SplitGenerator(
+                    name=datasets.Split.TEST, gen_kwargs={
+                        "transcript_path": transcripts["test"],
+                        "audio_archives": [dl_manager.iter_archive(archive) for archive in audio_archives["test"]],
+                        "local_extracted_archive": local_extracted_archives.get("test"),
+                    }
+                ),
+                ]
+        print(train_splits)
         return train_splits + [
             datasets.SplitGenerator(
                 name=datasets.Split.TEST, gen_kwargs={
@@ -133,6 +142,7 @@ class Aloresb(datasets.GeneratorBasedBuilder):
         Generate examples as dicts.
         """
         transcripts = {}
+        print('XXXXX', transcript_path)
         with open(transcript_path, encoding="utf-8") as f:
             for line in f:
                 audio_id, text = line.strip().split("\t")

audio/wolof/dev/transcripts.txt

@@ -1,4 +1,3 @@
-
 WOL_03_lect_0001	 seen suuf dafa nangu lu ñu fa ji mu sax
 WOL_03_lect_0003	 war na joxe kayit yii yépp bala weer wu dee
 WOL_03_lect_0004	 ndaxam mu ngi ne cocc ak téeméeri atam