audioset.py

# coding=utf-8

"""AudioSet sound event classification dataset."""


import os
import json
import textwrap
import datasets
import itertools
import typing as tp
import pandas as pd
from pathlib import Path
from huggingface_hub import hf_hub_download


SAMPLE_RATE = 32_000

_HOMEPAGE = "https://huggingface.co/datasets/confit/audioset"

_BALANCED_TRAIN_FILENAME = 'balanced_train_segments.zip'
_EVAL_FILENAME = 'eval_segments.zip'

ID2LABEL = json.load(
    open(hf_hub_download("huggingface/label-files", "audioset-id2label.json", repo_type="dataset"), "r")
)
LABEL2ID = {v:k for k, v in ID2LABEL.items()}
CLASSES = list(set(LABEL2ID.keys()))


class AudioSetConfig(datasets.BuilderConfig):
    """BuilderConfig for AudioSet."""
    
    def __init__(self, features, **kwargs):
        super(AudioSetConfig, self).__init__(version=datasets.Version("0.0.1", ""), **kwargs)
        self.features = features


class AudioSet(datasets.GeneratorBasedBuilder):

    BUILDER_CONFIGS = [
        AudioSetConfig(
            features=datasets.Features(
                {
                    "file": datasets.Value("string"),
                    "audio": datasets.Audio(sampling_rate=SAMPLE_RATE),
                    "sound": datasets.Sequence(datasets.Value("string")), 
                    "label": datasets.Sequence(datasets.features.ClassLabel(names=CLASSES)), 
                }
            ),
            name="balanced", 
            description="",
        ), 
    ]

    DEFAULT_CONFIG_NAME = "balanced"

    def _info(self):
        return datasets.DatasetInfo(
            description="",
            features=self.config.features,
            supervised_keys=None,
            homepage="",
            citation="",
            task_templates=None,
        )

    def _preprocess_metadata_csv(self, csv_file):
        df = pd.read_csv(csv_file, skiprows=2, sep=', ', engine='python')
        df.rename(columns={'positive_labels': 'ids'}, inplace=True)
        df['ids'] = [label.strip('\"').split(',') for label in df['ids']]
        df['filename'] = (
            'Y' + df['# YTID'] + '.wav'
        )
        return df[['filename', 'ids']]

    def _split_generators(self, dl_manager):
        """Returns SplitGenerators."""
        if self.config.name == 'balanced':
            archive_path = dl_manager.extract(_BALANCED_TRAIN_FILENAME)
        elif self.config.name == 'unbalanced':
            archive_path = dl_manager.extract(_UNBALANCED_TRAIN_FILENAME)
        test_archive_path = dl_manager.extract(_EVAL_FILENAME)
        
        return [
            datasets.SplitGenerator(
                name=datasets.Split.TRAIN, gen_kwargs={"archive_path": archive_path, "split": "train"}
            ), 
            datasets.SplitGenerator(
                name=datasets.Split.TEST, gen_kwargs={"archive_path": test_archive_path, "split": "test"}
            ), 
        ]

    def _generate_examples(self, archive_path, split=None):
        extensions = ['.wav']

        if split == 'train':
            if self.config.name == 'balanced':
                train_metadata_csv = f"{_HOMEPAGE}/resolve/main/metadata/balanced_train_segments.csv"
            elif self.config.name == 'unbalanced':
                train_metadata_csv = f"{_HOMEPAGE}/resolve/main/metadata/unbalanced_train_segments.csv"
            metadata_df = self._preprocess_metadata_csv(train_metadata_csv) # ['filename', 'ids']
        elif split == 'test':
            test_metadata_csv = f"{_HOMEPAGE}/resolve/main/metadata/eval_segments.csv"
            metadata_df = self._preprocess_metadata_csv(test_metadata_csv) # ['filename', 'ids']
        
        class_labels_indices_df = pd.read_csv(
            f"{_HOMEPAGE}/resolve/main/metadata/class_labels_indices.csv"
        ) # ['index', 'mid', 'display_name']
        mid2label = {
            row['mid']:row['display_name'] for idx, row in class_labels_indices_df.iterrows()
        }

        def default_find_classes(audio_path):
            fileid = Path(audio_path).name
            ids = metadata_df.query(f'filename=="{fileid}"')['ids'].values.tolist()
            ids = [
                mid2label.get(mid, None) for mid in flatten(ids)
            ]
            return ids

        _, _walker = fast_scandir(archive_path, extensions, recursive=True)

        for guid, audio_path in enumerate(_walker):
            yield guid, {
                "id": str(guid),
                "file": audio_path, 
                "audio": audio_path, 
                "sound": default_find_classes(audio_path), 
                "label": default_find_classes(audio_path), 
            }


def flatten(list2d):
    return list(itertools.chain.from_iterable(list2d))


def fast_scandir(path: str, exts: tp.List[str], recursive: bool = False):
    # Scan files recursively faster than glob
    # From github.com/drscotthawley/aeiou/blob/main/aeiou/core.py
    subfolders, files = [], []

    try:  # hope to avoid 'permission denied' by this try
        for f in os.scandir(path):
            try:  # 'hope to avoid too many levels of symbolic links' error
                if f.is_dir():
                    subfolders.append(f.path)
                elif f.is_file():
                    if os.path.splitext(f.name)[1].lower() in exts:
                        files.append(f.path)
            except Exception:
                pass
    except Exception:
        pass

    if recursive:
        for path in list(subfolders):
            sf, f = fast_scandir(path, exts, recursive=recursive)
            subfolders.extend(sf)
            files.extend(f)  # type: ignore

    return subfolders, files