Stereo Tool
https://www.forums.stereotool.com/

Is artificial intelligence and machine learning also relevant for the sound processors of the future?
https://www.forums.stereotool.com/viewtopic.php?t=21658		 Page 2 of 2
Author:  near05 [ Fri Oct 20, 2017 7:55 pm ]
Post subject:  Re: Is artificial intelligence and machine learning also relevant for the sound processors of the future?
Quote:
I actually like what Auto EQ does with voice right now.... are you hearing weird things?
In my opinion, there is much more you can get out of AutoEQ. Please take a look at the following plugin.

http://www.sonible.com/blog/introducing-smart-eq-plus/

I tried that once, and the results are impressive. The EQ analyzes the content and automatically filters out frequencies. Furthermore, it is still possible to readjust, but this is hardly necessary. They rely on these techniques. Don't you think you could benefit from that, too?
Author:  Bojcha [ Sat Oct 21, 2017 1:19 am ]
Post subject:  Re: Is artificial intelligence and machine learning also relevant for the sound processors of the future?
Quote:
@near05: For Natural Dynamics it would help (because you really don't want ND to be active on that). For all the other things, not.

Having said that, I would rather improve the ND algorithm to not respond to speech (which should be possible, and it already doesn't do much with it) than make it depend on some detection that may or may not fail. There are other sounds that you also don't want ND to act on, and if there's music with vocals you also don't want ND to do anything to the vocals.
--------------------------
That would be good! ND is working on speech too much. That's one of reasons why i keep it on moderate level.
Maybe simple speech/voice detector is simplest solution for that. However full dynamics detector is needed for music too. I really don't want ABBA track to be changed by ND, but currently much is.
Author:  Renox [ Tue Oct 24, 2017 9:11 pm ]
Post subject:  Re: Is artificial intelligence and machine learning also relevant for the sound processors of the future?
Hi,

I just came across this topic. I can only tell about my experience with Auphonic (https://auphonic.com/audio_examples) which is an automated web service for post production which is quite often used by Podcasters.
As far as i know they use machine learning to optimize their algorithms. I had very good results with bad recordings, which are often the results when trying to record a speaker in a lecture hall at the university. Especially the ease of use and their integration with external services makes it a tool which is worth to try out.
Page 2 of 2 All times are UTC+02:00
Powered by phpBB® Forum Software © phpBB Limited
https://www.phpbb.com/