The MUSAN (Multi Speaker ANnotated) dataset is used for speech, music, and noise classification.
speech → konuşma
music → müzik
noise → gürültü
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"
