I'd like to find an app that can learn each voice in a household, and then track the percentage/average time, relative to each other, per time frame that each voice speaks. It seems like this should be fairly simple, but I can't seem to find an app that does this.
- Feb 6, 2017
- 96,816
- 12,644
- 113
Hi, I haven't come across a app like that, maybe some other members can chime in.
I'll leave a Link to register so you can communicate here as a guest account you can only post questions but can't reply
https://forums.androidcentral.com/showthread.php?t=409154
I'll leave a Link to register so you can communicate here as a guest account you can only post questions but can't reply
https://forums.androidcentral.com/showthread.php?t=409154
VidJunky
Well-known member
- Dec 6, 2011
- 6,232
- 1,230
- 113
A quick search shows that it may be possible but a few sources say it isn't easy and others reference doing this to musical tracks. I doubt you'll find this even in desktop form and based on the difficulty level expressed in the article I looked at probably not at all on Android.
Not sure what you're goal would be with this but if it is important to you I believe you could do something similar but do the work your self. So no automatic way.
If the parties aren't talking over one another a simple audio app that shows soundwaves and is multi tracked could do this. You'd simply segment out each persons audio and move it to its own track. Then at the end of the recording you could see who spoke most by the length of their track. Then a little non-racist math would give you the percentage of each speaker. This method would also be good for excluding silent times or give the silent times their own track to judge the percentage of non-dialoged time.
Without revealing anything personal or private could you possibly explain or give an example of why or what the goal is?
Not sure what you're goal would be with this but if it is important to you I believe you could do something similar but do the work your self. So no automatic way.
If the parties aren't talking over one another a simple audio app that shows soundwaves and is multi tracked could do this. You'd simply segment out each persons audio and move it to its own track. Then at the end of the recording you could see who spoke most by the length of their track. Then a little non-racist math would give you the percentage of each speaker. This method would also be good for excluding silent times or give the silent times their own track to judge the percentage of non-dialoged time.
Without revealing anything personal or private could you possibly explain or give an example of why or what the goal is?
Similar threads
Facebook
Twitter
Instagram