Please declare the dataset license (CC0-1.0?)
#3
by pj1garcia - opened
Thanks for maintaining this parquet mirror — it's the only CV17 source
that still works with datasets 5.0 now that loader scripts are gone.
The repo card doesn't declare a license, and the README doesn't state
whether this is a faithful copy of the upstream corpus. Mozilla releases
Common Voice under CC0-1.0, but that doesn't carry over automatically
without a declaration here.
I'm building an assistive-technology product that will be sold to state
vocational rehabilitation agencies, so I need a clear license on anything
I derive artifacts from. Would you be able to add license: cc0-1.0 to
the dataset card, and note whether this is a full mirror or a filtered
subset?
pj1garcia changed discussion status to closed
pj1garcia changed discussion status to open