Sonarworks SoundID VoiceAI - First attempt!

[HCMarkus]

Those singers are pros, for sure, but if you've got a good ear, decent tone, and some tuning software, you can do great BVs in the studio, too! I often find myself filling in BVs for artists with whom I work... it is fun finding parts and crafting a sweet BV blend with one's voice.

And let me add... tuning BVs is not always required. Disparity lends richness, as long as it is close enough to proper pitch to avoid havoc! Maybe tuning one part of a doubled harmony, which is actually a great way to make one part into its own double.

 

[Nekujak]

Those singers are pros, for sure, but if you've got a good ear, decent tone, and some tuning software, you can do great BVs in the studio, too! I often find myself filling in BVs for artists with whom I work... it is fun finding parts and crafting a sweet BV blend with one's voice.

And let me add... tuning BVs is not always required. Disparity lends richness, as long as it is close enough to proper pitch to avoid havoc! Maybe tuning one part of a doubled harmony, which is actually a great way to make one part into its own double.

Oh for sure. I learned how to produce big vocals from several pro engineers many years ago, and for some styles, it's practically a requirement. But for more intimate singer-songwriter stuff, a wall of background vocals isn't always appropriate, so that just leaves my feeble lead vocal exposed to the world

 

[Mike Greene]

Well sadly, I must admit that D is my own inadequate vocal - but I agree with you 100% (I always hated the way it finished weakly on the word "dead", among other flaws). I never professed to be a singer, even though I've inflicted my talentless voice on many a bewildered audience, and thankfully, never released this particular track commercially

Believe me, I've gottan an earful of critiques from my wife every time I try to sing something, so I've developed a thick skin to the point if someone happens to compliment my singing, I never believe them

Wow, that surprises me, although listening back, I may have been a little harsh, since D doesn't sound bad at all. (I, along with most peopole with decent hearing, would prefer your singing over mine. ) I just figured it was one of the derivatives because it seemed to have less "interpretation" in the performance, and seemed a little "straight." I was basically trying to outsmart the game, as opposed to making an assessment of what sounds good.

What's interesting, though, is that I'd totally believe A or B, at least for that section. Kinda scary.

 

[CrimsonWarlock]

Well, I'm one of those who think VoiceAI is good enough for my backing vocals. But in my style of music (progrock and its neighbors) the vocals, including the lead, are more back in the mix than in pop so we can get away with more so to speak. Besides that, in this genre the backing vocals are more used as an instrument, probably why vocoders are also still popular in this realm (I totally love vocoders)

I've got the pop-voices expansion for Christmas, but I actually wanted the rock-voices (for obvious reasons), so my wife made good on that this last weekend for my birthday. I'm having loads of fun with this stuff.

 

[Skymuse]

Just wanted to post an update on my journey so far with SoundID VoiceAI...

At present, I've finished producing 6 songs of my current 10-song pop project, and up to this point, have decided not to use VoiceAI at all, not even for background vocals. This was partly a functional decision - VoiceAI is not quite good enough yet to carry a convincing lead vocal for an entire song, and isn't really necessary for imparting character and variety to backgrounds; and partly a moral choice - I wanted to avoid having to explain the use of AI in my productions, even if it's in an assistive capacity, so before resorting to any kind of AI help, prefer to use human contributions.

So... I hired a singer to record vocals on songs where my wife's voice wasn't a good fit. Don't worry, this was at my wife's suggestion, so all is still well at home . When needed, I've applied some formant shifting and EQ effects to alter the character of a vocal, but for the most part, have stuck with the singers' natural voices.

During the writing process though, I have used VoiceAI to transform my quick rough vocal sketches into a female voice to hear how it will sound. This has actually been quite helpful in zeroing-in on a sound or style that I'm after, and for creating quick demos that the singers can work from.


I'm actually not opposed to using VoiceAI to transform lead and background vocals in final productions. It's just that for this particular project, given some of its limitations and the current attitudes surrounding AI, I'm choosing to sidestep it altogether. It's possible I might choose to use VoiceAI in the future.

By the way, it looks like Sonarworks has been trying to make their rather ugly VoiceAI web pages slightly more useful. They now have a page where you can audition all the voices they offer, both in the core product and all expansions.

Has anyone else been using VoiceAI, ReSing, Ace Studio, etc. and have anything interesting to report?

I just downloaded VoiceAI and ReSing to test and created four alternate vocals - one female in VoiceAI, and 2 female + 1 male in ReSing. I'm wrestling with timeline issues, just a workflow problem in Pro Tools from what I can tell, so no good demos yet.

The VoiceID female singer has a thinness and depth that I actually like. The three ReSing voices have a "sameness" quality. I haven't decided yet if this is bad or good; all are following the same guide track so they should sound similar.

Also, I somewhat prefer how VoiceID feels more lightweight and snappier running on a M2 Max Mac Studio. For the current intro price I may buy the perpetual version and call this my AI experimenting tool. As others have noted, this technology is in its infancy. If voice replacement fits my workflow I can imagine buying a pricier professional product in the reasonable future.

 

[Skymuse]

Well sadly, I must admit that D is my own inadequate vocal - but I agree with you 100% (I always hated the way it finished weakly on the word "dead", among other flaws). I never professed to be a singer, even though I've inflicted my talentless voice on many a bewildered audience, and thankfully, never released this particular track commercially

Believe me, I've gottan an earful of critiques from my wife every time I try to sing something, so I've developed a thick skin to the point if someone happens to compliment my singing, I never believe them

Actually, I find it kind of encouraging that VoiceAI was able to impart some modest improvement. It gives me hope that I might actually be able to use VoiceAI to shore up some of my other deficient vocals.

Your assessments are all spot on, including your comments about Tyler's vocal

I spotted D as your original right away and thought A might be as well. D sounds most human, especially the trailing last word. I think your voice sounds strong enough that you can probably address issues like the weak finish through comping takes and also using compression / saturation to build density. I also like your original vocal in your first post.

 

[HCMarkus]

I think your voice sounds strong enough

+1.

Don't sell yourself short.

 

[Nekujak]

Wow, that surprises me, although listening back, I may have been a little harsh, since D doesn't sound bad at all. (I, along with most peopole with decent hearing, would prefer your singing over mine. )

I think your voice sounds strong enough that you can probably address issues like the weak finish through comping takes and also using compression / saturation to build density. I also like your original vocal in your first post.

Don't sell yourself short.

I appreciate the words of encouragement

 

[Nekujak]

Thought I'd share a couple of interesting little VoiceAI examples. As I mentioned earlier, I sometimes use VoiceAI while working out a song so I can hear what it will sound like with a female vocal.

Here are short excerpts from two pop tracks I've been working on. These demo vocals were quickly laid down by me (no tuning, no cleanup, no comping) and then run thru VoiceID:

Ferocious:
View attachment Nekujak - Ferocious DEMO.mp3

Storm:
View attachment Nekujak - Storm DEMO.mp3

The only vocal processing is EQ, compression, and saturation, which is applied to the VoiceAI output. The source vocal remained completely raw and unprocessed. There's some high frequency phaseyness, but I believe that's mostly due to the stereo saturator, and can be tamed.

I think the results are actually pretty good - no weird accents, mispronunciations, or strange vocal artifacts. Considering it's a female vocal derived from a male source, I think it's pretty impressive.

Clearly, this is a specialized type of vocal. As we already know, most of VoiceAI's weaknesses become evident when a more natural vocal sound is called for.

 

[Mike Greene]

Those are both very impressive, although I especially like Ferocious. I'd hit the Like button on that one if it appeared in my Spotify rotation.

 

[Nekujak]

Those are both very impressive, although I especially like Ferocious. I'd hit the Like button on that one if it appeared in my Spotify rotation.

Wow, thanks Mike. I guess I'm gonna to have to work faster to get these tracks finished