Aynursusuz/musan-audio-dataset

MUSAN Audio Dataset

The MUSAN (Multi Speaker ANnotated) dataset is used for speech, music, and noise classification.

CLASSES

• speech → konuşma
• music → müzik
• noise → gürültü

USING

from datasets import load_dataset

ds = load_dataset("Aynursusuz/musan-audio-dataset")
print(ds["train"][0]["audio"])
print(ds["train"][0]["label"])  # "speech", "music", "noise"