News for hackers. Stuff that matters.

A comment on Mozilla Common Voice Dataset: More data, more languages. Parent comment

Re:Mozilla Common Voice Dataset: More data, more languages by indogooner on Wednesday July 01, 2020 @04:48AM (#23699323)
Curious to know why don't researchers use Audiobooks/Videos and transcript when data is not available? Is it because these do not capture different dialects/accents?
[ Parent | on HN ]
Re:Mozilla Common Voice Dataset: More data, more languages by bluGill on Wednesday July 01, 2020 @11:10AM (#23702263)
Those tend to be read in artificial voices. They are useful for some things, but it isn't real speach so it isn't as generally useful.
[ Parent | on HN ]
Re:Mozilla Common Voice Dataset: More data, more languages by petargyurov on Wednesday July 01, 2020 @05:26AM (#23699574)
I suppose there might some copyright issues with such content (?)
[ Parent | on HN ]
Re:Mozilla Common Voice Dataset: More data, more languages by mjepronk on Wednesday July 01, 2020 @06:12AM (#23699852)
Well there is Librivox...
[ Parent | on HN ]
Re:Mozilla Common Voice Dataset: More data, more languages by est31 on Wednesday July 01, 2020 @07:24AM (#23700256)
Indeed, that's what one of the famous established datasets, LibriSpeech, bases on.
[ Parent | on HN ]