rir-noise.py

# coding=utf-8

"""RIR-Noise dataset."""


import os
import textwrap
import datasets
import itertools
import typing as tp
from pathlib import Path


SAMPLE_RATE = 16_000

_RIR_NOISE_URL = 'https://www.openslr.org/resources/28/rirs_noises.zip'
_AUDIO_TYPES = ['pointsource_noises', 'real_rirs_isotropic_noises', 'simulated_rirs']


class RIRNoiseConfig(datasets.BuilderConfig):
    """BuilderConfig for RIR-Noise."""
    
    def __init__(self, features, **kwargs):
        super(RIRNoiseConfig, self).__init__(version=datasets.Version("0.0.1", ""), **kwargs)
        self.features = features


class RIRNoise(datasets.GeneratorBasedBuilder):

    BUILDER_CONFIGS = [
        RIRNoiseConfig(
            features=datasets.Features(
                {
                    "file": datasets.Value("string"),
                    "audio": datasets.Audio(sampling_rate=SAMPLE_RATE),
                    "label": datasets.ClassLabel(names=_AUDIO_TYPES),
                }
            ),
            name="rir-noise", 
            description=textwrap.dedent(
                """\
                A database of simulated and real room impulse responses, isotropic and point-source noises. 
                """
            ),
        ), 
    ]

    def _info(self):
        return datasets.DatasetInfo(
            description="A database of simulated and real room impulse responses, isotropic and point-source noises.",
            features=self.config.features,
            supervised_keys=None,
            homepage="https://www.openslr.org/28",
            citation="""
                @inproceedings{ko2017study,
                  title={A study on data augmentation of reverberant speech for robust speech recognition},
                  author={Ko, Tom and Peddinti, Vijayaditya and Povey, Daniel and Seltzer, Michael L and Khudanpur, Sanjeev},
                  booktitle={2017 IEEE international conference on acoustics, speech and signal processing (ICASSP)},
                  pages={5220--5224},
                  year={2017},
                  organization={IEEE}
                }
            """,
            task_templates=None,
        )

    def _split_generators(self, dl_manager):
        """Returns SplitGenerators."""
        archive_path = dl_manager.download_and_extract(_RIR_NOISE_URL)
        return [
            datasets.SplitGenerator(
                name=datasets.Split.TRAIN, gen_kwargs={"archive_path": archive_path, "split": "train"}
            ), 
        ]

    def _generate_examples(self, archive_path, split=None):
        extensions = ['.wav']
        _, _walker = fast_scandir(archive_path, extensions, recursive=True)

        if split == 'train':
            _walker = [fileid for fileid in _walker]

        for guid, audio_path in enumerate(_walker):
            if 'pointsource_noises' in audio_path:
                label = 'pointsource_noises'
            elif 'real_rirs_isotropic_noises' in audio_path:
                label = 'real_rirs_isotropic_noises'
            elif 'simulated_rirs' in audio_path:
                label = 'simulated_rirs'
            yield guid, {
                "id": str(guid),
                "file": audio_path, 
                "audio": audio_path, 
                "label": label, 
            }


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