Create musan.py

musan.py

@@ -0,0 +1,115 @@
+# coding=utf-8
+
+"""MUSAN dataset."""
+
+
+import os
+import textwrap
+import datasets
+import itertools
+import typing as tp
+from pathlib import Path
+
+
+SAMPLE_RATE = 16_000
+
+_MUSAN_URL = 'https://www.openslr.org/resources/17/musan.tar.gz'
+_AUDIO_TYPES = ['music', 'noise', 'speech']
+
+
+class MusanConfig(datasets.BuilderConfig):
+    """BuilderConfig for MUSAN."""
+    
+    def __init__(self, features, **kwargs):
+        super(MusanConfig, self).__init__(version=datasets.Version("0.0.1", ""), **kwargs)
+        self.features = features
+
+
+class MUSAN(datasets.GeneratorBasedBuilder):
+
+    BUILDER_CONFIGS = [
+        MusanConfig(
+            features=datasets.Features(
+                {
+                    "file": datasets.Value("string"),
+                    "audio": datasets.Audio(sampling_rate=SAMPLE_RATE),
+                    "label": datasets.ClassLabel(names=_AUDIO_TYPES),
+                }
+            ),
+            name="gtzan", 
+            description=textwrap.dedent(
+                """\
+                MUSAN is a corpus of music, speech, and noise recordings.
+                """
+            ),
+        ), 
+    ]
+
+    def _info(self):
+        return datasets.DatasetInfo(
+            description="MUSAN is a corpus of music, speech, and noise recordings.",
+            features=self.config.features,
+            supervised_keys=None,
+            homepage="https://www.openslr.org/17",
+            citation="""
+                @misc{musan2015,
+                  author = {David Snyder and Guoguo Chen and Daniel Povey},
+                  title = {{MUSAN}: {A} {M}usic, {S}peech, and {N}oise {C}orpus},
+                  year = {2015},
+                  eprint = {1510.08484},
+                  note = {arXiv:1510.08484v1}
+                }
+            """,
+            task_templates=None,
+        )
+
+    def _split_generators(self, dl_manager):
+        """Returns SplitGenerators."""
+        archive_path = dl_manager.download_and_extract(_MUSAN_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):
+            yield guid, {
+                "id": str(guid),
+                "file": audio_path, 
+                "audio": audio_path, 
+                "label": Path(audio_path).parent.parent.stem, 
+            }
+
+
+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