Upload dataset.py

dataset.py

@@ -8,11 +8,11 @@ _DESCRIPTION = """\
 Bu dataset mp3 formatdagi audio fayllar va tsv metadata fayllardan iborat.
 Audio fayllar .tar arxiv ichida saqlangan va tsv faylda fayl nomlari (masalan, H3H38EY38D8.mp3) keltirilgan.
 """
-
 _HOMEPAGE = "https://huggingface.co/datasets/Elyordev/new_dataset_stt"
-_LICENSE = "MIT"  # O'zingizga mos litsenziya
+_LICENSE = "MIT"
 
 # Har bir split uchun .tsv va .tar fayllarning repo ichidagi joylashuvi.
+# (Sizning repo tarkibingizda train/train.tsv, train/train.tar, va hokazo bo'lsa)
 _URLS = {
     "train": {
         "tsv": "train/train.tsv",
@@ -28,10 +28,12 @@ _URLS = {
     },
 }
 
+
 class MyDatasetSTTConfig(datasets.BuilderConfig):
     def __init__(self, **kwargs):
         super(MyDatasetSTTConfig, self).__init__(**kwargs)
 
+
 class MyDatasetSTT(datasets.GeneratorBasedBuilder):
     VERSION = datasets.Version("1.0.0")
     BUILDER_CONFIGS = [
@@ -63,7 +65,11 @@ class MyDatasetSTT(datasets.GeneratorBasedBuilder):
         )
 
     def _split_generators(self, dl_manager):
-        # Har bir split uchun .tsv va .tar fayllarni dl_manager yordamida yuklab olamiz.
+        """
+        Har bir split uchun .tsv va .tar fayllarni dl_manager yordamida yuklab olamiz.
+        Orqa fon’da tar-faylni to‘liq extract qilish shart emas.
+        'tar://...' URI orqali audio oqimini bevosita o‘qish mumkin.
+        """
         downloaded_files = {}
         for split in _URLS:
             downloaded_files[split] = {
@@ -96,23 +102,27 @@ class MyDatasetSTT(datasets.GeneratorBasedBuilder):
 
     def _generate_examples(self, tsv_path: str, tar_path: str) -> Iterator[Tuple[int, dict]]:
         """
-        Har bir .tsv fayldagi qatordan misol yield qilamiz.
+        Har bir .tsv fayldagi qatordan misol (example) yaratamiz.
         Audio faylga murojaat qilish uchun "tar://" sintaksisidan foydalanamiz:
         Bu format: "tar://<tar fayl yo'li>#<tsv fayldagi path>".
         """
         with open(tsv_path, encoding="utf-8") as f:
             reader = csv.DictReader(f, delimiter="\t")
             for idx, row in enumerate(reader):
+                # MP3 fayl nomi, masalan "H3H38EY38D8.mp3"
+                mp3_file = row["path"]
+
                 # Audio fayl uchun URI: masalan, "tar://.../train.tar#H3H38EY38D8.mp3"
-                audio_ref = f"tar://{tar_path}#{row['path']}"
+                audio_ref = f"tar://{tar_path}#{mp3_file}"
+
                 yield idx, {
                     "id": row["id"],
-                    "path": row["path"],
+                    "path": mp3_file,
                     "sentence": row["sentence"],
                     "duration": float(row.get("duration", 0.0)),
                     "age": row.get("age", ""),
                     "gender": row.get("gender", ""),
                     "accents": row.get("accents", ""),
                     "locale": row.get("locale", ""),
-                    "audio": audio_ref,
+                    "audio": audio_ref,  # tar ichidan oqish
                 }