Update librispeech_asr.py

librispeech_asr.py

@@ -113,59 +113,50 @@ class LibrispeechASR(datasets.GeneratorBasedBuilder):
         # (Optional) In non-streaming mode, we can extract the archive locally to have actual local audio files:
         local_extracted_archive = dl_manager.extract(archive_path) if not dl_manager.is_streaming else {}
 
+        splits = []
         if self.config.name == "clean":
-            dev_splits = [
+            splits.extend([
                 datasets.SplitGenerator(
-                    name=datasets.Split.VALIDATION,
+                    name="validation",  # Changed from datasets.Split.VALIDATION
                     gen_kwargs={
                         "local_extracted_archive": local_extracted_archive.get("dev"),
                         "files": dl_manager.iter_archive(archive_path["dev"]),
                     },
-                )
-            ]
-            test_splits = [
+                ),
                 datasets.SplitGenerator(
-                    name=datasets.Split.TEST,
+                    name="test",  # Changed from datasets.Split.TEST
                     gen_kwargs={
                         "local_extracted_archive": local_extracted_archive.get("test"),
                         "files": dl_manager.iter_archive(archive_path["test"]),
                     },
                 )
-            ]
-            return dev_splits + test_splits
-            
+            ])
         elif self.config.name == "other":
-            train_splits = [
+            splits.extend([
                 datasets.SplitGenerator(
                     name="train.500",
                     gen_kwargs={
                         "local_extracted_archive": local_extracted_archive.get("train.500"),
                         "files": dl_manager.iter_archive(archive_path["train.500"]),
                     },
-                )
-            ]
-            dev_splits = [
+                ),
                 datasets.SplitGenerator(
-                    name=datasets.Split.VALIDATION,
+                    name="validation",
                     gen_kwargs={
                         "local_extracted_archive": local_extracted_archive.get("dev"),
                         "files": dl_manager.iter_archive(archive_path["dev"]),
                     },
-                )
-            ]
-            test_splits = [
+                ),
                 datasets.SplitGenerator(
-                    name=datasets.Split.TEST,
+                    name="test",
                     gen_kwargs={
                         "local_extracted_archive": local_extracted_archive.get("test"),
                         "files": dl_manager.iter_archive(archive_path["test"]),
                     },
                 )
-            ]
-            return train_splits + dev_splits + test_splits
-            
+            ])
         elif self.config.name == "all":
-            train_splits = [
+            splits.extend([
                 datasets.SplitGenerator(
                     name="train.clean.100",
                     gen_kwargs={
@@ -187,8 +178,6 @@ class LibrispeechASR(datasets.GeneratorBasedBuilder):
                         "files": dl_manager.iter_archive(archive_path["train.other.500"]),
                     },
                 ),
-            ]
-            dev_splits = [
                 datasets.SplitGenerator(
                     name="validation.clean",
                     gen_kwargs={
@@ -203,8 +192,6 @@ class LibrispeechASR(datasets.GeneratorBasedBuilder):
                         "files": dl_manager.iter_archive(archive_path["dev.other"]),
                     },
                 ),
-            ]
-            test_splits = [
                 datasets.SplitGenerator(
                     name="test.clean",
                     gen_kwargs={
@@ -219,8 +206,9 @@ class LibrispeechASR(datasets.GeneratorBasedBuilder):
                         "files": dl_manager.iter_archive(archive_path["test.other"]),
                     },
                 ),
-            ]
-            return train_splits + dev_splits + test_splits
+            ])
+
+        return splits
 
     def _generate_examples(self, files, local_extracted_archive):
         """Generate examples from a LibriSpeech archive_path."""