dvoice-darija3

darija/texts/test.tsv

@@ -1,4 +1,4 @@
-wav	words	duration
+path	words	duration
 clipping-bbc22c9d-37c7-4426-8069-e759bd81480f.wav	ممكن تعيطي ليها في لبورطابل	2.2
 frequency_dropout-735439b0-9a4c-4cfb-9f64-0707c18e8a20.wav	صور هديه جميله جدا	3.272666666666667
 time_dropout-38132f4d-c788-4e0e-aaf9-e24ad60638ab.wav	دراغون بول	2.6006666666666667

darija/texts/train.tsv

@@ -1,4 +1,4 @@
-wav	words	duration
+path	words	duration
 clipping-306514b0-d498-41ff-b6ef-333cc36045d8.wav	النار سمعت واحد قال لصاحبه رخص يكون عندنا شويه التراب	5.934172335600908
 1ddb113f-bbbd-426b-be1f-0e0be670ae4d.wav	مرام راكان اسمع والدلائل شيء واحد	4.704580498866213
 time_dropout-1e3fcef4-25ec-4548-8075-7868e5b307bb.wav	ممكن نكون جاي في الطريق	3.2783900226757368

dvoice-darija.py

@@ -0,0 +1,85 @@
+import csv
+import os
+import json
+
+import datasets
+from datasets.utils.py_utils import size_str
+from tqdm import tqdm
+
+from .languages import LANGUAGES
+from .release_stats import STATS
+
+
+
+# TODO: change "streaming" to "main" after merge!
+_BASE_URL = "https://huggingface.co/datasets/makiour/dvoice-Darija/tree/main/darija/"
+
+_AUDIO_URL = _BASE_URL + "wavs/{path}.wav"
+
+_TRANSCRIPT_URL = _BASE_URL + "texts/{split}.tsv"
+
+_N_SHARDS_URL = _BASE_URL + "n_shards.json"
+
+
+class CommonVoiceConfig(datasets.BuilderConfig):
+    """BuilderConfig for CommonVoice."""
+
+    def __init__(self, name, version, **kwargs):
+        self.language = kwargs.pop("language", None)
+        self.release_date = kwargs.pop("release_date", None)
+        self.num_clips = kwargs.pop("num_clips", None)
+        self.num_speakers = kwargs.pop("num_speakers", None)
+        self.validated_hr = kwargs.pop("validated_hr", None)
+        self.total_hr = kwargs.pop("total_hr", None)
+        self.size_bytes = kwargs.pop("size_bytes", None)
+        self.size_human = size_str(self.size_bytes)
+        description = (
+           "Finetune Whisper with Dvoice Darija Dataset"
+        )
+        super(CommonVoiceConfig, self).__init__(
+            name=name,
+            version=datasets.Version(version),
+            description=description,
+            **kwargs,
+        )
+
+
+class CommonVoice(datasets.GeneratorBasedBuilder):
+    DEFAULT_WRITER_BATCH_SIZE = 1000
+
+    BUILDER_CONFIGS = [
+        CommonVoiceConfig(
+            name=lang,
+            version=STATS["version"],
+            language=LANGUAGES[lang],
+            release_date=STATS["date"],
+            num_clips=lang_stats["clips"],
+            num_speakers=lang_stats["users"],
+            validated_hr=float(lang_stats["validHrs"]) if lang_stats["validHrs"] else None,
+            total_hr=float(lang_stats["totalHrs"]) if lang_stats["totalHrs"] else None,
+            size_bytes=int(lang_stats["size"]) if lang_stats["size"] else None,
+        )
+        for lang, lang_stats in STATS["locales"].items()
+    ]
+
+    def _info(self):
+        total_languages = len(STATS["locales"])
+        total_valid_hours = STATS["totalValidHrs"]
+        description = (
+           "Finetune Whisper with Dvoice Darija Dataset"
+        )
+        features = datasets.Features(
+            {
+                "path": datasets.Value("string"),
+                "audio": datasets.features.Audio(sampling_rate=48_000),
+                "words": datasets.Value("string"),
+                "duration": datasets.Value("int64"),
+            }
+        )
+
+        return datasets.DatasetInfo(
+            description=description,
+            features=features,
+            supervised_keys=None,
+            version=self.config.version,
+        )