Yield full path to audio files

ksponspeech.py

@@ -15,9 +15,9 @@
 """The Korean Spontaneous Speech Corpus for Automatic Speech Recognition (KsponSpeech)"""
 
 
-import os
-import datasets
+from os import path
 
+import datasets
 
 _CITATION = """\
 @article{bang2020ksponspeech,
@@ -46,9 +46,7 @@ class KsponSpeech(datasets.GeneratorBasedBuilder):
 
     @property
     def manual_download_instructions(self):
-        return (
-            "To use KsponSpeech, data files must be downloaded manually to a local drive. Please submit your request on the official website (https://aihub.or.kr/aidata/105). Once your request is approved, download all files, extract .zip files in one folder, and load the dataset with `datasets.load_dataset('ksponspeech', data_dir='path/to/folder')`."
-        )
+        return "To use KsponSpeech, data files must be downloaded manually to a local drive. Please submit your request on the official website (https://aihub.or.kr/aidata/105). Once your request is approved, download all files, extract .zip files in one folder, and load the dataset with `datasets.load_dataset('ksponspeech', data_dir='path/to/folder')`."
 
     def _info(self):
         return datasets.DatasetInfo(
@@ -66,12 +64,12 @@ class KsponSpeech(datasets.GeneratorBasedBuilder):
 
     def _split_generators(self, dl_manager):
         """Returns SplitGenerators."""
-        data_dir = os.path.abspath(os.path.expanduser(dl_manager.manual_dir))
+        self.data_dir = path.abspath(path.expanduser(dl_manager.manual_dir))
         return [
             datasets.SplitGenerator(
                 name=datasets.Split.TRAIN,
                 gen_kwargs={
-                    "filepath": os.path.join(data_dir, "scripts/train.trn"),
+                    "filepath": path.join(self.data_dir, "scripts/train.trn"),
                     "split": "train",
                 },
             ),
@@ -79,30 +77,32 @@ class KsponSpeech(datasets.GeneratorBasedBuilder):
                 name=datasets.Split.TEST,
                 gen_kwargs={
                     "filepath": {
-                        "clean": os.path.join(data_dir, "scripts/eval_clean.trn"),
-                        "other": os.path.join(data_dir, "scripts/eval_other.trn"),
+                        "clean": path.join(self.data_dir, "scripts/eval_clean.trn"),
+                        "other": path.join(self.data_dir, "scripts/eval_other.trn"),
                     },
-                    "split": "test"
+                    "split": "test",
                 },
             ),
             datasets.SplitGenerator(
                 name=datasets.Split.VALIDATION,
                 gen_kwargs={
-                    "filepath": os.path.join(data_dir, "scripts/dev.trn"),
-                    "split": "dev",
+                    "filepath": path.join(self.data_dir, "scripts/dev.trn"),
+                    "split": "validation",
                 },
             ),
         ]
 
     def _generate_examples(self, filepath, split):
-        """ Yields examples as (key, example) tuples. """
+        """Yields examples as (key, example) tuples."""
         if split is "test":
-            with open(filepath["clean"], encoding="utf-8") as f1, open(filepath["other"], encoding="utf-8") as f2:
+            with open(filepath["clean"], encoding="utf-8") as f1, open(
+                filepath["other"], encoding="utf-8"
+            ) as f2:
                 data = "\n".join([f1.read().strip(), f2.read().strip()])
                 for id_, row in enumerate(data.split("\n")):
                     path, sentence = tuple(row.split(" :: "))
                     yield id_, {
-                        "path": path,
+                        "file": path.join(self.data_dir, path),
                         "sentence": sentence,
                     }
         else:
@@ -111,6 +111,6 @@ class KsponSpeech(datasets.GeneratorBasedBuilder):
                 for id_, row in enumerate(data.split("\n")):
                     path, sentence = tuple(row.split(" :: "))
                     yield id_, {
-                        "path": path,
-                        "sentence": sentence
+                        "file": path.join(self.data_dir, path),
+                        "sentence": sentence,
                     }