Upload vietnamese-speaker-video.py

vietnamese-speaker-video.py

@@ -0,0 +1,131 @@
+# Copyright 2023 Thinh T. Duong
+import os
+import datasets
+from huggingface_hub import HfFileSystem
+from typing import List, Tuple
+
+
+logger = datasets.logging.get_logger(__name__)
+fs = HfFileSystem()
+
+
+_CITATION = """
+
+"""
+_DESCRIPTION = """
+    This dataset contain videos of Vietnamese speakers.
+"""
+_HOMEPAGE = "https://github.com/tanthinhdt/vietnamese-av-asr"
+_REPO_PATH = "datasets/phdkhanh2507/vietnamese-speaker-video"
+_REPO_URL = f"https://huggingface.co/{_REPO_PATH}/resolve/main"
+_URLS = {
+    "meta": f"{_REPO_URL}/metadata/" + "{channel}.parquet",
+    "video": f"{_REPO_URL}/video/" + "{channel}.zip",
+}
+_CONFIGS = ["all"]
+if fs.exists(_REPO_PATH + "/metadata"):
+    _CONFIGS.extend([
+        os.path.basename(file_name)[:-8]
+        for file_name in fs.listdir(_REPO_PATH + "/metadata", detail=False)
+        if file_name.endswith(".parquet")
+    ])
+
+
+class VietnameseSpeakerVideoConfig(datasets.BuilderConfig):
+    """Vietnamese Speaker Video configuration."""
+
+    def __init__(self, name, **kwargs) -> None:
+        """
+        :param name:    Name of subset.
+        :param kwargs:  Arguments.
+        """
+        super().__init__(
+            name=name,
+            version=datasets.Version("1.0.0"),
+            description=_DESCRIPTION,
+            **kwargs,
+        )
+
+
+class VietnameseSpeakerVideo(datasets.GeneratorBasedBuilder):
+    """Vietnamese Speaker Video dataset."""
+
+    BUILDER_CONFIGS = [VietnameseSpeakerVideoConfig(name) for name in _CONFIGS]
+    DEFAULT_CONFIG_NAME = "all"
+
+    def _info(self) -> datasets.DatasetInfo:
+        features = datasets.Features({
+            "id": datasets.Value("string"),
+            "channel": datasets.Value("string"),
+            "video": datasets.Value("string"),
+            "fps": datasets.Value("int8"),
+            "sampling_rate": datasets.Value("int64"),
+        })
+
+        return datasets.DatasetInfo(
+            description=_DESCRIPTION,
+            features=features,
+            homepage=_HOMEPAGE,
+            citation=_CITATION,
+        )
+
+    def _split_generators(
+        self, dl_manager: datasets.DownloadManager
+    ) -> List[datasets.SplitGenerator]:
+        """
+        Get splits.
+        :param dl_manager:  Download manager.
+        :return:            Splits.
+        """
+        config_names = _CONFIGS[1:] if self.config.name == "all" else [self.config.name]
+
+        metadata_paths = dl_manager.download(
+            [_URLS["meta"].format(channel=channel) for channel in config_names]
+        )
+        video_dirs = dl_manager.download_and_extract(
+            [_URLS["video"].format(channel=channel) for channel in config_names]
+        )
+
+        video_dict = {
+            channel: video_dir for channel, video_dir in zip(config_names, video_dirs)
+        }
+
+        return [
+            datasets.SplitGenerator(
+                name=datasets.Split.TRAIN,
+                gen_kwargs={
+                    "metadata_paths": metadata_paths,
+                    "video_dict": video_dict,
+                },
+            ),
+        ]
+
+    def _generate_examples(
+        self, metadata_paths: List[str],
+        video_dict: dict,
+    ) -> Tuple[int, dict]:
+        """
+        Generate examples from metadata.
+        :param metadata_paths:      Paths to metadata.
+        :param video_dict:          Paths to directory containing videos.
+        :yield:                     Example.
+        """
+        dataset = datasets.load_dataset(
+            "parquet",
+            data_files=metadata_paths,
+            split="train",
+            use_auth_token=True,
+        )
+        for i, sample in enumerate(dataset):
+            channel = sample["channel"]
+            video_path = os.path.join(
+                video_dict[channel], channel, sample["id"] + ".avi"
+            )
+
+            yield i, {
+                "id": sample["id"],
+                "channel": channel,
+                "video": video_path,
+                "fps": sample["fps"],
+                "sampling_rate": sample["sampling_rate"],
+            }