dummy url for test download

Uodate docstring

fix sampling rate

* add sample option

missed to remove one printing

remove printing to stdout

add loading bar with tqdm

votes to int64

add sampling_rate 48k to match common voice

remove sampling rate from feature

add more clean log

avoid dicts in dataset

fix touple error

fix dict type and handle empty values

remove skips

add one more

add count of annotations

add bad annotations

add bad annotations

skip whole speaker who causes problems

add list of "bad" files

readd logs

catch error loading file

add temp log

check that "val_recordings" exists

fix typo in split name

add key "channel"

all is keyword, cant be used in SplitGenerator

add loading script, dataset info

Upload dataset_infos.json

Upload data/train-00000-of-00001.parquet with git-lfs

Create README.md

Update README.md

initial commit