3 ms·
Surely they (Berkeley) could use some speech-to-text software to do this relatively well, easy, and free. There's also bound to be people who would have volunte
by k-mcgrady 10y ago
Surely they (Berkeley) could use some speech-to-text software to do this relatively well, easy, and free. There's also bound to be people who would have volunteered to do it for free so they didn't lose access.
- PeterisP 10y agoNo, that's not considered sufficient. The Department found that [...] many had automatic captioning generated by YouTube’s speech recognition technology. In March 2015, the Department selected 30 videos. [..] for review. Automatically generated captions were inaccurate and incomplete, making the content inaccessible to individuals with hearing disabilities.
- k-mcgrady 10y agoI thought that would be the case. What I really meant was for Berkley to play the content while speech-to-text is running and then check it for errors before making real (not live) captions/transcripts available.