Dreamtronics has introduced Vocoflex a real-time voice creation and transformation plugin.
Vocoflex lets you change voice gender, blend male and female voices, create backing vocals, create original voice and more.
Features:
- Define a target voice either by importing a voice sample or using randomization.
- Transform a voice sample or mic input into a target voice in real-time.
- Transform and morph between multiple target voices.
- Add movement to any voice transformation using your DAW’s automation.
- Control voices via MIDI by setting up waypoints and assigning them to your MIDI controller.
Vocoflex Walkthrough:
To prevent unauthorized usage, Vocoflex embeds an inaudible audio watermark into its output. Also as a precautionary measure, Vocoflex requires ID verification, to help curb the usage of voice samples without a singer’s consent.
Pricing and Availability:
Vocoflex is available now with an intro price of $159.20, normally $199.
Timmmberrrrrr
Finally I can be a lumberjack (with consent).
Cool processor.
That’s actually very cool but creepy at the same time…
I’m sure Dreamtronics will be fine with everyone using this for training purposes without paying them.
it;s still a long way from sounding like a human vocal but neat synthesis
Clever, but my second impression was wondering how long it will take for hackers to break through the ID & watermark aspects. Vocal AI shenanigans are already being sharpened for casual thievery.
My first impression: how about a sample of lawyers rubbing their hands together at high speed, eager to fight over the inevitable misuses. I’m waiting to see when or if this proves as useful as the numerous other ways of treating vocals. I’ll try to hold back on my skepticism until the concept solidifies more.
No time at all, ‘AI watermarking’ is complete BS and the people hiding behind know so. Virtually all scams this might be used for are conducted over the phone – who’s going to check phone conversations for watermarks? Grandma? The phone company? It’s just a liability deflection. Few if any scam victims are clued in enough to even record a call, and even then I’ll bet it’s trivially easy to find and defeat the watermarking process, which has to be stealthy enough not to distract from legitimate use, and robust enough to survive heavy bandpass filtering and compression (over phone lines) or studio processing (for music). It is good that they’re addressing the issue up front, but the reality is that this tech is staggeringly easy to abuse and disclaimers seem to have been designed by a lawyer rather than a technologist.
From a purely artistic standpoint this product seems OK, but looking at the pricing you can tell they hope they’ve got the next autotune on their hands. In an increasingly crowded marketplace for voice synthesis tech, with open-source options also becoming available. I’m not sure what the musical use cases are; maybe cheaper backing vocals, or ‘vocal repair’ when the lead singer gets trashed the night before the session. But it seems to me that lead singers won’t want to be associated with this product at all, and backing singers will hate it because it threatens to put them out of a job. It could end up as a curiosity plugin for experimental sound transformation, like a vocoder on steroids…but then they’re competing with Autotune and all of the many sound mangling plugins/techniques.
I can see bigger applications in the film/video world, where you might want drastic vocal transformations or to overcome difficulties from poor recording/mic placement/performance. Automatic dialog replacement is expensive and tedious for everyone involved, so reducing it is a big win for post-production houses. I think they’d be wiserto redo it as a pro-tools plugin, jack the price up to $500+ and focus on this market segment, where there’s a very clear value-for-money proposition.
there is no demo version, is there? (probably because of that ID verification thing?) that’s a pity.. the technology looks promising but at that price point i’d need to test it myself..