add documentation on download size and streaming

README.md

@@ -130,7 +130,7 @@ dataloader = DataLoader(afrispeech, batch_sampler=batch_sampler)
 from datasets import load_dataset
 from torch.utils.data import DataLoader
 
-afrispeech = load_dataset("tobiolatunji/afrispeech-200", "isizulu", split="train")
+afrispeech = load_dataset("tobiolatunji/afrispeech-200", "isizulu", split="train", streaming=True)
 dataloader = DataLoader(afrispeech, batch_size=32)
 ```