pianos.py

# coding=utf-8

"""Pianos sound classification dataset."""


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

from ._pianos import _PITCHES

_COMPRESSED_FILENAME = 'pianos.zip'

_NAMES = [
    "PearlRiver",
    "YoungChang",
    "Steinway-T",
    "Hsinghai",
    "Kawai",
    "Steinway",
    "Kawai-G",
    "Yamaha",
]

SAMPLE_RATE = 16_000

_CITATION = """\
@dataset{zhaorui_liu_2021_5676893,
  author       = {Zhaorui Liu, Monan Zhou, Shenyang Xu, Yuan Wang, Zhaowen Wang, Wei Li and Zijin Li},
  title        = {CCMUSIC DATABASE: A Music Data Sharing Platform for Computational Musicology Research},
  month        = {nov},
  year         = {2021},
  publisher    = {Zenodo},
  version      = {1.1},
  doi          = {10.5281/zenodo.5676893},
  url          = {https://doi.org/10.5281/zenodo.5676893}
}
"""

_DESCRIPTION = """\
Piano-Sound-Quality is a dataset of piano sound. It consists of 8 kinds of pianos including PearlRiver, YoungChang, Steinway-T, Hsinghai, Kawai, Steinway, Kawai-G, Yamaha(recorded by Shaohua Ji with SONY PCM-D100). Data was annotated by students from the China Conservatory of Music (CCMUSIC) in Beijing and collected by Monan Zhou.
"""


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


class Pianos(datasets.GeneratorBasedBuilder):
    
    BUILDER_CONFIGS = [
        PianosConfig(
            features=datasets.Features(
                {
                    "file": datasets.Value("string"),
                    "audio": datasets.Audio(sampling_rate=SAMPLE_RATE),
                    "sound": datasets.Value("string"),
                    "label": datasets.ClassLabel(names=_NAMES), 
                }
            ),
            name="pianos", 
            description=textwrap.dedent(_DESCRIPTION), 
        ), 
    ]

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

    def _split_generators(self, dl_manager):
        data_files = dl_manager.extract(_COMPRESSED_FILENAME)
        dataset = []
        
        for path in dl_manager.iter_files([data_files]):
            fname = os.path.basename(path)
            if fname.endswith(".wav"):
                dataset.append(
                    {
                        "file": path, 
                        "audio": path, 
                        "label": os.path.basename(os.path.dirname(path)),
                        "sound": os.path.basename(os.path.dirname(path)),
                    }
                )

        random.shuffle(dataset)
        count = len(dataset)
        p80 = int(0.8 * count)
        p90 = int(0.9 * count)

        return [
            datasets.SplitGenerator(
                name=datasets.Split.TRAIN, gen_kwargs={"files": dataset[:p80]}
            ),
            datasets.SplitGenerator(
                name=datasets.Split.VALIDATION, gen_kwargs={"files": dataset[p80:p90]}
            ),
            datasets.SplitGenerator(
                name=datasets.Split.TEST, gen_kwargs={"files": dataset[p90:]}
            ),
        ]

    def _generate_examples(self, files):
        for guid, path in enumerate(files):
            yield guid, {
                "id": str(guid), 
                "file": path["file"], 
                "audio": path["audio"],
                "label": path["label"],
                "sound": path["sound"],
            }