Datasets:

---

mozilla-foundation

/

common_voice_17_0

like 333

Follow

Mozilla Foundation 299

When running the example in readme

from datasets import load_dataset
cv_17 = load_dataset("mozilla-foundation/common_voice_17_0", "hi", split="train", streaming=True)

print(next(iter(cv_17)))

I get the error below. I am running datasets 3.6.0 and huggingface hub 0.28.1, and python 3.12. It is related to the audio column, because if I remove that before iterating it works. I have also tried a different machine. And also different dataset (gigaspeech), which worked. So I suspect this is related to the dataset.

---------------------------------------------------------------------------
CastError                                 Traceback (most recent call last)
Cell In[2], line 4
      1 from datasets import load_dataset
      2 cv_17 = load_dataset("mozilla-foundation/common_voice_17_0", "hi", split="train", streaming=True)
----> 4 print(next(iter(cv_17)))

File ~/workdir/tts/mimi/.venv/lib/python3.12/site-packages/datasets/iterable_dataset.py:2270, in IterableDataset.__iter__(self)
   2267         yield formatter.format_row(pa_table)
   2268     return
-> 2270 for key, example in ex_iterable:
   2271     # no need to format thanks to FormattedExamplesIterable
   2272     yield example

File ~/workdir/tts/mimi/.venv/lib/python3.12/site-packages/datasets/iterable_dataset.py:1856, in FormattedExamplesIterable.__iter__(self)
   1849     formatter = get_formatter(
   1850         self.formatting.format_type,
   1851         features=self._features if not self.ex_iterable.is_typed else None,
   1852         token_per_repo_id=self.token_per_repo_id,
   1853     )
   1854 if self.ex_iterable.iter_arrow:
   1855     # feature casting (inc column addition) handled within self._iter_arrow()
-> 1856     for key, pa_table in self._iter_arrow():
   1857         batch = formatter.format_batch(pa_table)
   1858         for example in _batch_to_examples(batch):

File ~/workdir/tts/mimi/.venv/lib/python3.12/site-packages/datasets/iterable_dataset.py:1888, in FormattedExamplesIterable._iter_arrow(self)
   1886         pa_table = pa_table.append_column(column_name, col)
   1887 if pa_table.schema != schema:
-> 1888     pa_table = cast_table_to_features(pa_table, self.features)
   1889 yield key, pa_table

File ~/workdir/tts/mimi/.venv/lib/python3.12/site-packages/datasets/table.py:2215, in cast_table_to_features(table, features)
   2203 """Cast a table to the arrow schema that corresponds to the requested features.
   2204 
   2205 Args:
   (...)   2212     table (`pyarrow.Table`): the casted table
   2213 """
   2214 if sorted(table.column_names) != sorted(features):
-> 2215     raise CastError(
   2216         f"Couldn't cast\n{_short_str(table.schema)}\nto\n{_short_str(features)}\nbecause column names don't match",
   2217         table_column_names=table.column_names,
   2218         requested_column_names=list(features),
   2219     )
   2220 arrays = [cast_array_to_feature(table[name], feature) for name, feature in features.items()]
   2221 return pa.Table.from_arrays(arrays, schema=features.arrow_schema)

CastError: Couldn't cast
client_id: string
path: string
sentence_id: string
sentence: string
sentence_domain: string
up_votes: string
down_votes: string
age: string
gender: string
variant: string
locale: string
segment: string
accent: string
audio: struct<bytes: binary, path: string>
  child 0, bytes: binary
  child 1, path: string
to
{'client_id': Value(dtype='string', id=None), 'path': Value(dtype='string', id=None), 'audio': Audio(sampling_rate=48000, mono=True, decode=True, id=None), 'sentence': Value(dtype='string', id=None), 'up_votes': Value(dtype='int64', id=None), 'down_votes': Value(dtype='int64', id=None), 'age': Value(dtype='string', id=None), 'gender': Value(dtype='string', id=None), 'accent': Value(dtype='string', id=None), 'locale': Value(dtype='string', id=None), 'segment': Value(dtype='string', id=None), 'variant': Value(dtype='string', id=None)}
because column names don't match