I would like to axe a question about the state of amplifier profiling as it relates to what appears to be a rushing train towards us of Vocal profiling

[bookumdano4]

Cool long title huh? In a really old Craig archive from another century, I asked how long till we all have a black box that lets you input a mic sound of your voice and at the output you get, say Frank Sinatra. The current state of amplifier profiling makes me wonder if we're maybe within 15 years of dead-accurate vocal profiling software.

 

English is hard. Vocaloid can't really do it naturally.... software that adds breaths etc isn't really magic.... samples of vocal phrases are, well, samples of vocal phrases.

 

In my increasing smiles over amp profiling, I realize it's not really a sampling process. Or maybe it is. But whatever, a lot of it is sounding pretty cool.

 

While I would agree there will only ever be one Frank Sinatra, Robert Plant, or Tiny Tim, it would SEEM to me that if the software gets where it needs to be, all any of us would really have to do is sit there for an hour with our mic/vocal profiling software, and a copy of a Sinatra record.... and just monkey around with inflections/timing at the mic input to eventually fairly adequately nail a pretty good representation of the Sinatra presets...which would no doubt be known as "Voice Alotsa" or "FrankieCroon" or other non-trademark-infringing refs like we currently have for amp cloning.

 

Amp profiling is pretty cool. I think about all the stuff I do with my fingers on the guitar and how realistic the profiling... or even sims.... are as a result of that. Especially the profiling part.

 

So...what's the status of this cutting edge technology as it may pertain to lead vocals? 15 more years until Vocal profiling? Six months?

[Anderton]

Very interesting question. I really don't know the answer...there are so many elements that make up what makes a voice distinctive. It's not just about formants and spectral distribution, but phrasing and such.

 

I do know that I can model Craig Anderton's voice perfectly, but I guess that's not particularly impressive

[UstadKhanAli]

We don't have vocal modeling in that sense just yet. But thankfully, we have AutoTune, which when used on stun makes all our voices sound far more similar to each other!

[Phil O'Keefe]

I think your 15 years estimate is far closer to reality than your six month estimate. That's just my gut feeling. Vocals are exceptionally complex…