my_stt_dataset.py

import os
import csv
import datasets
from datasets import Audio, BuilderConfig

# Konfiguratsiya sinfi: datasetning til qisqartmasi va ma'lumotlar joylashgan papkani belgilaydi.
class STTConfig(BuilderConfig):
    def __init__(self, language_abbr, data_dir, **kwargs):
        """
        Args:
            language_abbr (str): Til qisqartmasi, masalan "uz".
            data_dir (str): Dataset joylashgan asosiy papka, masalan "Dataset_STT".
            **kwargs: Qolgan BuilderConfig parametrlar.
        """
        super().__init__(**kwargs)
        self.language_abbr = language_abbr
        self.data_dir = data_dir

# Dataset yuklash skripti
class MySTTDataset(datasets.GeneratorBasedBuilder):
    """
    Uzbek STT dataset yuklash skripti.
      - Audio fayllar .tar arxiv ichida saqlanadi.
      - Transkripsiya ma'lumotlari TSV faylda mavjud.
      - "audio" ustuni Audio() tipida aniqlanib, HF Dataset Viewer tomonidan "play" tugmasi ko'rsatiladi.
    """
    VERSION = datasets.Version("1.0.0")

    # Faqat bitta konfiguratsiya ishlatilmoqda.
    BUILDER_CONFIGS = [
        STTConfig(
            name="uz",
            version=datasets.Version("1.0.0"),
            description="Uzbek subset of the STT dataset",
            language_abbr="uz",
            data_dir="Dataset_STT",  # Bu yerda asosiy papka nomi
        )
    ]
    DEFAULT_CONFIG_NAME = "uz"

    def _info(self):
        """
        Datasetning ustunlari aniqlanadi.
          - "audio" ustuni Audio() tipida, bu orqali fayl avtomatik dekodlanadi.
        """
        return datasets.DatasetInfo(
            description="Uzbek STT dataset: audio fayllar .tar arxivda, transcriptions esa TSV faylda saqlanadi.",
            features=datasets.Features({
                "id": datasets.Value("string"),
                "audio": Audio(sampling_rate=None),  # asl sampling rate saqlanadi
                "sentence": datasets.Value("string"),
                "duration": datasets.Value("float"),
                "age": datasets.Value("string"),
                "gender": datasets.Value("string"),
                "accents": datasets.Value("string"),
                "locale": datasets.Value("string"),
            }),
            supervised_keys=None,
            version=self.VERSION,
        )

    def _split_generators(self, dl_manager):
        """
        Har bir split (train, test, validation) uchun:
          - Tar arxiv va mos TSV fayllarning yo'llari aniqlanadi.
          - dl_manager.extract() yordamida tar fayllar extract qilinadi.
        """
        config = self.config  # STTConfig obyekti
        base_dir = config.data_dir  # Masalan: "Dataset_STT"
        lang = config.language_abbr   # Masalan: "uz"

        # Audio va transkript fayllarining yo'llarini aniqlaymiz:
        train_tar = os.path.join(base_dir, "audio", lang, "train.tar")
        train_tsv = os.path.join(base_dir, "transcript", lang, "train.tsv")

        test_tar = os.path.join(base_dir, "audio", lang, "test.tar")
        test_tsv = os.path.join(base_dir, "transcript", lang, "test.tsv")

        val_tar = os.path.join(base_dir, "audio", lang, "validation.tar")
        val_tsv = os.path.join(base_dir, "transcript", lang, "validation.tsv")

        # Tar arxiv extract qilinadi:
        train_tar_extracted = dl_manager.extract(train_tar)
        test_tar_extracted = dl_manager.extract(test_tar)
        val_tar_extracted = dl_manager.extract(val_tar)

        return [
            datasets.SplitGenerator(
                name=datasets.Split.TRAIN,
                gen_kwargs={
                    "archive_dir": train_tar_extracted,  # Extract qilingan papka
                    "tsv_path": train_tsv,
                },
            ),
            datasets.SplitGenerator(
                name=datasets.Split.TEST,
                gen_kwargs={
                    "archive_dir": test_tar_extracted,
                    "tsv_path": test_tsv,
                },
            ),
            datasets.SplitGenerator(
                name=datasets.Split.VALIDATION,
                gen_kwargs={
                    "archive_dir": val_tar_extracted,
                    "tsv_path": val_tsv,
                },
            ),
        ]

    def _generate_examples(self, archive_dir, tsv_path):
        """
        TSV faylini qatorma-qator o'qib, metadata lug'atini yaratadi va
        extract qilingan archive papkasidan mos .mp3 faylni topadi.
        
        Eslatma: Audio ustunini faqat fayl yo'li sifatida uzatsangiz,
        HF Dataset Viewer bu ustunni audio sifatida ko'rsata olmaydi.
        Shu sababli, audio ustuni quyidagicha dictionary shaklida bo'lishi kerak:
            {"path": mp3_path, "bytes": <audio_file_baytlari>}
        """
        with open(tsv_path, "r", encoding="utf-8") as f:
            reader = csv.DictReader(f, delimiter="\t")
            for idx, row in enumerate(reader):
                audio_id = row["id"]
                mp3_file = audio_id + ".mp3"
                mp3_path = os.path.join(archive_dir, mp3_file)

                if os.path.isfile(mp3_path):
                    # Audio faylni baytlar shaklida o'qib olamiz
                    with open(mp3_path, "rb") as audio_file:
                        audio_bytes = audio_file.read()
                    yield idx, {
                        "id": audio_id,
                        "audio": {"path": mp3_path, "bytes": audio_bytes},
                        "sentence": row.get("sentence", ""),
                        "duration": float(row.get("duration", 0.0)),
                        "age": row.get("age", ""),
                        "gender": row.get("gender", ""),
                        "accents": row.get("accents", ""),
                        "locale": row.get("locale", ""),
                    }
                else:
                    # Agar mos audio fayl topilmasa, ushbu yozuvni o'tkazib yuboramiz.
                    continue