Upload indo-gs2.py with huggingface_hub

indo-gs2.py

@@ -0,0 +1,98 @@
+import csv
+import os
+import zipfile
+from typing import List
+
+import datasets
+
+_DATASETNAME = "id_asr"
+
+_DESCRIPTION = """\
+This dataset contains transcribed audio data for Indonesian. The dataset consists of audio files and a CSV file.
+The CSV file contains the audio ID and transcription of the audio in the file.
+"""
+_LANGUAGES = ["ind"]
+
+_SOURCE_VERSION = "1.0.0"
+
+class IdASR(datasets.GeneratorBasedBuilder):
+    """Indonesian ASR training dataset."""
+
+    VERSION = datasets.Version(_SOURCE_VERSION)
+
+    BUILDER_CONFIGS = [
+        datasets.BuilderConfig(
+            name="default",
+            version=VERSION,
+            description="Indonesian ASR dataset",
+        )
+    ]
+
+    DEFAULT_CONFIG_NAME = "default"
+
+    def _info(self) -> datasets.DatasetInfo:
+        features = datasets.Features(
+            {
+                "id": datasets.Value("string"),
+                "path": datasets.Value("string"),
+                "audio": datasets.Audio(sampling_rate=16_000),
+                "sentence": datasets.Value("string"),
+            }
+        )
+        return datasets.DatasetInfo(
+            description=_DESCRIPTION,
+            features=features
+        )
+
+    def _split_generators(self, dl_manager: datasets.DownloadManager) -> List[datasets.SplitGenerator]:
+        manual_dir = dl_manager.manual_dir
+
+        return [
+            datasets.SplitGenerator(
+                name=datasets.Split.TRAIN,
+                gen_kwargs={"archive_path": os.path.join(manual_dir, "train.zip")},
+            ),
+            datasets.SplitGenerator(
+                name=datasets.Split.VALIDATION,
+                gen_kwargs={"archive_path": os.path.join(manual_dir, "val.zip")},
+            ),
+            datasets.SplitGenerator(
+                name=datasets.Split.TEST,
+                gen_kwargs={"archive_path": os.path.join(manual_dir, "test.zip")},
+            ),
+        ]
+
+    def _generate_examples(self, archive_path: str):
+        extract_dir = archive_path.rstrip(".zip")
+
+        with zipfile.ZipFile(archive_path, "r") as zip_ref:
+            zip_ref.extractall(extract_dir)
+
+        csv_path = None
+        for root, _, files in os.walk(extract_dir):
+            for file in files:
+                if file.endswith(".csv"):
+                    csv_path = os.path.join(root, file)
+                    break
+            if csv_path:
+                break
+
+        if not csv_path:
+            raise FileNotFoundError("CSV file not found inside the archive.")
+
+        with open(csv_path, "r", encoding="utf-8") as f:
+            reader = csv.DictReader(f)
+            for row in reader:
+                audio_id = row["id"]
+                relative_path = row["path"]
+                sentence = row["sentence"]
+
+                # Construct full path to audio file
+                full_audio_path = os.path.join(extract_dir, relative_path)
+
+                yield audio_id, {
+                    "id": audio_id,
+                    "path": full_audio_path,
+                    "audio": full_audio_path,
+                    "sentence": sentence,
+                }