File size: 4,146 Bytes
6109566 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 |
# 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"],
}
|