Refactor

test.py

@@ -86,19 +86,20 @@ class TestASR(datasets.GeneratorBasedBuilder):
         )
 
     def _split_generators(self, dl_manager):
-        audio_path = dl_manager.download(_DATA_URL)
-        local_extracted_archive = dl_manager.extract(audio_path) if not dl_manager.is_streaming else None
+        # audio_path = dl_manager.download(_DATA_URL)
+        # local_extracted_archive = dl_manager.extract(audio_path) if not dl_manager.is_streaming else None
         meta_path = dl_manager.download(_PROMPTS_URLS)
+        audio_path = dl_manager.download_and_extract(_DATA_URL)
         return [datasets.SplitGenerator(
             name=datasets.Split.TEST,
             gen_kwargs={
                 "meta_path": meta_path["test"],
-                "audio_files": dl_manager.iter_archive(audio_path),
-                "local_extracted_archive": local_extracted_archive,
+                "audio_files": dl_manager.iter_files(audio_path),
+                # "local_extracted_archive": local_extracted_archive,
             }
         )]
 
-    def _generate_examples(self, meta_path, audio_files, local_extracted_archive):
+    def _generate_examples(self, meta_path, audio_files):#, local_extracted_archive):
         """Lorem ipsum."""
         data_fields = list(self._info().features.keys())
         metadata = {}
@@ -114,11 +115,11 @@ class TestASR(datasets.GeneratorBasedBuilder):
                                       "ngram": ngram}
 
         id_ = 0
-        for path, f in audio_files:
+        for path in audio_files:
             print(path, f)
-            _, audio_name = os.path.split(path)
+            *_, audio_name = os.path.split(path)
             if audio_name in metadata:
                 result = dict(metadata[audio_name])
-                result["audio"] = {"path": path, "bytes":f.read()}
+                result["audio"] = path#{"path": path, "bytes":f.read()}
                 yield id_, result
                 id_ +=1