RE: [BeyondBitcoin RSVP] Whaletank Hangout #228 | 2017-10-14 | Sat @230PM UTC

You are viewing a single comment's thread from:

RE: [BeyondBitcoin RSVP] Whaletank Hangout #228 | 2017-10-14 | Sat @230PM UTC | Over 650 $BD in Awards!

alexpmorris (65)in #eos-project • 7 years ago

srt2vtt can now normalize captions via machine learning, as well as cut and/or splice video+audio+captions with the help of the excellent SubTitle Edit by adding actionable cues to the timeline.

7 years ago in #eos-project by alexpmorris (65)

Sort:

nutela (61) 7 years ago (edited)

So you send each cue one by one to the punctuator?

Sorry hadn't seen this.

$0.00

[-]

alexpmorris (65) 7 years ago

that's exactly what we talked about, that's the version I had sent you, thought you had even tried it already!

$0.00

[-]

nutela (61) 7 years ago

Alex,

When did you build it in?
Did Chucky tried it?
How does it work? Like you don't have to edit and make sentences on YouTube anymore?
So where does it fit in the work-flow? (YT auto captioning, srt2vtt per subtitle track? So before mixing? Remember we use tracks separated for each speaker. Also if you insert speaker name will it mess up the alligner if we want to cut out the audio too later?)

Let's have a meeting and have a constructive group-chat pr talk about this say on Discord or Mumble.

$0.00

[-]

nutela (61) 7 years ago

This is very cool Alex, if you could make a video on how it works in action that'll be even better. I tried running it but mono crashed.

$0.00