add Nsynth dataset

AudioLoader/Music.py

@@ -697,3 +697,105 @@ def available_groups(self, group):
 
 
 
+
+
+class Nsynth(Dataset):
+    """Dataset class for Nsynth dataset.
+    Args:
+        root (str or Path): Path to the directory where the dataset is found or downloaded.
+
+        split (str):
+            Choose different dataset splits such as ``"train"``,``"valid"`` or ``"test"``c. (default: ``"train"``).
+
+        download (bool, optional):
+            Whether to download the dataset if it is not found at root path. (default: ``False``).
+
+    """
+
+
+    def __init__(self, root, split = "train", download = False):
+
+        url = f"http://download.magenta.tensorflow.org/datasets/nsynth/nsynth-{split}.jsonwav.tar.gz"
+        # Getting audio path
+        archive_name = f'nsynth-{split}.jsonwav.tar.gz'
+        folder_name = 'Nsynth'
+        download_path = os.path.join(root, folder_name)
+        assert split.upper()=="TRAIN" or split.upper()=="TEST" or split.upper()=="VALID" , f"split={split} is not present in this dataset"
+        self._path = os.path.join(download_path, f"nsynth-{split}", "audio")
+
+        checksum_dict = {"test": "5e6f8719bf7e16ad0a00d518b78af77d", "train":"fde6665a93865503ba598b9fac388660", "valid":"87e94a00a19b6dbc99cf6d4c0c0cae87"}
+
+        if download:
+            #file exists and extracted
+            if os.path.isfile(os.path.join(download_path,archive_name)) and os.path.exists(self._path):
+                print(f"Dataset archive exists, all files are extracted. Using all file from {self._path} ")
+            #file exists but not extracted
+            if os.path.isfile(os.path.join(download_path,archive_name)) and not os.path.exists(self._path):
+                print(f"Dataset archive exists, extracting archive:{os.path.join(download_path,archive_name)}")
+                extract_archive(os.path.join(download_path, archive_name))
+                print(f"Using all file from {self._path} ")
+            #file not exist
+            elif not os.path.isfile(os.path.join(download_path,archive_name)):
+                if not os.path.exists(download_path):
+                    os.makedirs(download_path)
+                try:
+                    download_url(url, download_path, hash_value= checksum_dict[split], hash_type='md5')
+                    extract_archive(os.path.join(download_path, archive_name))
+                    print(f"All files are extracted. Using all file from {self._path} ")
+
+                except:
+                    raise Exception('Auto download fails. '+
+                                    'You may want to download it manually from:\n'+
+                                    url+ '\n' +
+                                    f'Then, put it inside {download_path}')   
+        else:   
+            #archive is downloaded and extracted
+            if os.path.isfile(os.path.join(download_path, archive_name)) and not os.path.exists(self._path):
+                print(f"Dataset archive exists, all files are extracted. Using all file from {self._path} ")
+            #archive is downloaded but not extracted
+            elif os.path.isfile(os.path.join(download_path, archive_name)) and not os.path.exists(self._path):
+                print(f'archive:{os.path.join(download_path, archive_name)} exists, extracting...')
+                extract_archive(os.path.join(download_path, archive_name))  
+                print(f"Using all file from {self._path} ")
+            else:
+                raise FileNotFoundError(f"Dataset not found at {self._path}, please specify the correct location or set `download=True`")
+
+        print(f'Using all data at {self._path}')
+        self._walker = glob.glob(f"{self._path}/*.wav")
+
+        #load the label file
+        label_path = os.path.join(download_path, f"nsynth-{split}/examples.json") 
+        self.labels = json.load(open(label_path,"r")) 
+
+    def __getitem__(self, n):
+        """Load the n-th sample from the dataset.
+        Args:
+            n (int): The index of the sample to be loaded
+        Returns:
+            tuple: ``(path, wavform, sr, note, note_str, instr., instr._str, pitch, vel, qualities, qual._str, instr._fam, instr._fam_str, instr._src, instr._src_str )``
+        """
+        file_path = self._walker[n]  
+        waveform, sample_rate = torchaudio.load(file_path)
+        feature_dict = self.labels[os.path.basename(file_path).split(".")[0]]
+
+        batch = {'path': file_path,
+                 'waveform': waveform,
+                 'sample_rate': sample_rate,
+                 'note': feature_dict['note'],
+                 "note_str":feature_dict['note_str'],
+                 "instrument":feature_dict['instrument'],
+                 "instrument_str":feature_dict['note_str'],
+                 "pitch":feature_dict['pitch'],
+                 "velocity":feature_dict['velocity'],
+                 "qualities":feature_dict['qualities'],
+                 "qualities_str":feature_dict['qualities_str'],
+                 "instrument_family":feature_dict['instrument_family'],
+                 "instrument_family_str":feature_dict['instrument_family_str'],
+                 "instrument_source":feature_dict['instrument_source'],
+                 "instrument_source_str":feature_dict['instrument_source_str']
+                }
+
+        return batch
+
+    def __len__(self) -> int:
+        return len(self._walker)