You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The reference describes bookmarks as "one of the most important features of Manim Voiceover". Trying to use them seems to require that I install OpenAI Whisper, to parse the audio stream into text.
Why can't I simply mark the bookmarks myself? After all, I have a better idea of what I'm saying than the computer does, and I have to listen to what I've recorded anyway.
Advantages:
Fewer dependencies; simplifying installation.
Potentially greater reliability? (I haven't been able to test this.)
It just feels philosophically cleaner.
How can the new feature be used?
I imagine an interface similar to the one I use when using the RecorderService. I would listen back to my recording, and hit "Enter" at the moment the bookmark should occur. The terminal would prompt me by displaying the text surrounding the bookmark, so I'm prepared to hit "Enter" when it's spoken aloud.
It might be possible to make a better interface, but this would suffice.
Additional comments
"Simplifying installation" is more important than it sounds. I've been having trouble getting the dependencies to function at all, so I'm unable to use bookmarks. Which is frustrating, since I don't really want the AI transcription I've been failing to install, and the only thing I need it for is something I'd rather do myself.
The text was updated successfully, but these errors were encountered:
Description of proposed feature
The reference describes bookmarks as "one of the most important features of Manim Voiceover". Trying to use them seems to require that I install OpenAI Whisper, to parse the audio stream into text.
Why can't I simply mark the bookmarks myself? After all, I have a better idea of what I'm saying than the computer does, and I have to listen to what I've recorded anyway.
Advantages:
How can the new feature be used?
I imagine an interface similar to the one I use when using the
RecorderService
. I would listen back to my recording, and hit "Enter" at the moment the bookmark should occur. The terminal would prompt me by displaying the text surrounding the bookmark, so I'm prepared to hit "Enter" when it's spoken aloud.It might be possible to make a better interface, but this would suffice.
Additional comments
"Simplifying installation" is more important than it sounds. I've been having trouble getting the dependencies to function at all, so I'm unable to use bookmarks. Which is frustrating, since I don't really want the AI transcription I've been failing to install, and the only thing I need it for is something I'd rather do myself.
The text was updated successfully, but these errors were encountered: