That sounds implausible to me. There needs to at least be some information somewhere about who is speaking when for anything like that to be possible, or the software needs to have a sort of voice recognition AI. I haven't heard of that myself.
Do not PM me asking for help with Aegisub, Automation, karaoke or anything like that, I won't reply.
Ask your questions in the public forums. Use the Search function, somebody probably asked the same question before.
All rights reserved to call you insults if you ask poorly thought out or redundant questions.
Experimenting is free and is a great way to learn things.