Announcing Vocoflex – Real-time Voice Morphing Plugin

Dreamtonics is excited to announce Vocoflex, a real-time voice morphing plugin currently in extensive development and testing phases.


A highly innovative project under experiment, Vocoflex allows the transformation of any singing voice into new, user-customized voices. Users can define the target voice profile either by importing a voice sample as brief as 10 seconds or using a feature that randomly generates voices based on user-specified gender and tone.


Vocoflex features a minimalist interface, utilizing a novel approach where voice profiles are mixed by manipulating a light probe in a manner reminiscent of mixing lights and shadows. This interface displays imported samples and generated voice profiles as curves and points, placed automatically where similar voices become close to each other, allowing for intuitive and effortless manipulation.


Intended for music production and live performance, Vocoflex can transform voices with dynamically changing timbres, while maintaining an adjustable audio latency of as low as 45 milliseconds.


Using Vocoflex with Synthesizer V Studio

In addition to transforming real vocals, Vocoflex also works in synergy with the plugin version of Synthesizer V Studio, Dreamtonics’ vocal synthesis software that generates singing voice given lyrics and melodies. By placing Vocoflex in the effect chain following Synthesizer V Studio Plugin, users can significantly expand the range of voice identities from a limited set of voice databases installed, enabling the creation of rich and diverse vocal arrangements.



Safeguarding Vocoflex Against Misuses

While Vocoflex is designed for creative applications, Dreamtonics understands the potential risks and ethical implications of its use beyond these intended purposes. To prevent unauthorized usage, Vocoflex is optimized for singing voice rather than spoken voice and incorporates an audio watermarking feature. Undetectable to the ear, this feature embeds a user-unique license ID into the voice, enabling Dreamtonics to identify the creator of the audio sample, deterring users from importing voice samples without the singer’s consent – a practice prohibited by our usage terms.



Our advanced watermarking technology, integrated directly into the voice generation model, is designed to be tamper-proof, thus ensuring the integrity of the program against attempts to bypass the watermarking. Additionally, it is resilient against various audio manipulations such as mixing with background music and lossy compression through a telephone-quality channel.


Dreamtronics adheres to ethical AI usage, never blurring any moral and legal boundaries, and encourages users to use Vocoflex to expand their creative possibilities with the singer’s full knowledge and transparency.



Beta Testing Program Invitation


Dreamtonics is currently conducting a beta testing program for Vocoflex. We are seeking artists interested in exploring the creative capabilities of Vocoflex and providing feedback on its performance, functionality and potential use cases.


To ensure testing is conducted in a safe and constructive manner, we are launching a moderated program with limited slots available. Interested individuals can apply through our open sign-up form. Successful applicants will be granted an evaluation copy of Vocoflex after completing identity verification.


* At this time, we have not determined a plan for a public release.