Datasets:

---

speechcolab

/

gigaspeech2

like 61

Follow

SpeechColab 56

Hi, seems I can't obtain the wav data for split of dev and test using streaming-style of load_datasets

dataset = load_dataset(
    "speechcolab/gigaspeech2",
    "id",
    split=SPLIT,
    streaming=True
)

for i, row in enumerate(dataset):
     print(row["wav"]) #expectedly returns wav data like in the viewer, but it emits None 

it works on train split, tho