Datasets:

PedroDKE
/

LibriS2S

Tasks:

Automatic Speech Recognition

Modalities:

Formats:

Languages:

Size:

ArXiv:

Tags:

Speech-to-Speech translation

Libraries:

License:

Dataset card Data Studio Files Files and versions

PedroDKE commited on Jun 9

Commit

d7c9a0e

·

1 Parent(s): 11eba43

testing of loading hs dataset

Files changed (1) hide show

test_hf_ds_loader.py +5 -2

test_hf_ds_loader.py CHANGED Viewed

@@ -1,5 +1,8 @@
 from datasets import load_dataset, Audio
-dataset = load_dataset("csv", data_files="alignments/all_de_en_alligned_cleaned.csv", split="train")
 dataset = dataset.cast_column("DE_audio", Audio())
-print(dataset[0]["DE_audio"])  # This should return a dict with sampling_rate + path

 from datasets import load_dataset, Audio
+dataset = load_dataset(
+    "PedroDKE/LibriS2S",
+    trust_remote_code=True,
+)
 dataset = dataset.cast_column("DE_audio", Audio())
+print(dataset["train"][0]["DE_audio"])  # This should return a dict with sampling_rate + path