When it comes to researching hard-to-find information, audio and video interviews can be a treasure trove. However, unlike an interview appearing in print, it's hard to search that material. I can't tell you how many times I've wanted to reference something I partially remembered from an on-camera interview, but I either a) couldn't remember which one it was, or b) couldn't find the specific information I was looking for in a long video.

We had a lot of success with our Muppet Show transcripts, and I think the results were worth it. Friend of the wiki David Beukema might be able to instantly recall which episode it was where Kermit said [whatever], but most fans can't do that. The TMS transcripts help readers find that content.

Anyway, I'd like to propose the idea of housing transcripts of interviews here that would help everyone find a) that thing in the back of your head you can't quite seem to find again, and b) content for interviews you might not have listened to all the way through. There are obviously hundreds over the years, but the frequency with which these niche anecdotes have become available to us in the past year alone have increased aggressively due to the pandemic. Speaking strictly for myself, I'm never going to have the time to listen to them all (for example, thanks to someone's edit summary here on the wiki, I found myself on a YouTube channel for a podcast started last October with tons of interviews I'd love to listen to, but I have nowhere near the time to listen to the podcasts created by our friends at ToughPigs, nevermind some new host on the scene). That considered, it triggers my mild OCD and FOMO to know that there are little nuggets of info out there that we're missing out on. Enter: transcripts.

Adobe Premiere recently introduced a cloud transcribing service that I've been using for work, and it's surprisingly accurate. Higher than 95% I'd say. Now, we've got a transcript policy we might want to look at, but I think we could work out a system to include these maybe with a disclaimer up top that states these transcripts were generated by an AI. They won't be perfect, but they'll still include a lot more information than we had access to before.

As I write this, I'm transcribing Joan Ganz Cooney's 4.5 hour interview with The Archive of American Television. I'll update here when it's done doing its thing. Thoughts on housing these here? —Scott (talk) 19:59, 26 May 2021 (UTC)

I think there could be some value in this. But also, that's gonna be a lot of stuff to pour through. - Shane (talk) 20:32, 26 May 2021 (UTC)

Agreed. There's so much. I'm a bit ashamed to admit it, but I've never listened to all of Cooney's TV Archive interview. To me, that's a lot to pour through. Ditto the other interviews done by them. But if there's a collection of transcripts we can CTRL-F our way through, it would at least make the pouring easier. —Scott (talk) 20:36, 26 May 2021 (UTC)

Just wanted to note that some YouTube channels already have auto-generated captions for their videos and those can be downloaded as time-indexed srt files at https://downsub.com/ ...which has been a useful site. It just helped me verify sources with timestamps for the Kissinger, Eastwood, and Hanks notes on Potential guest stars. —Scott (talk) 11:14, 12 June 2021 (UTC)