Imagine a future where AI can transform your singing in real-time, enabling you to sing in a completely different voice to your own (and in tune!)? Well, imagine no more. In fact, there's about 100 apps promising - and some delivering - exactly that, not to mention all the music production software that offers alternatives to Antares Auto-Tune, the leading creative and vocal-enhancing application. Some of these specialize in real-time processing, while others are focused on off-line manipulation of existing recordings.
Voicemod is a Spanish company that originally specialized in voice-altering effects for gaming and online chat boards and evolved its software to enable real-time AI-voice transform and synthetic voices. Voicemod’s speech-to-speech AI technology allows users to change their voice in real-time with the same ease as switching in-game skins. With more than 45 million downloads and millions of users, the Voicemod app for Windows and Mac offers hundreds of AI voice options and a soundboard of over 150,000 sounds.
From tools used in gaming, streaming, content creation or simply for socializing online, Voicemod now adapted its technology to create an AI singing voice generator (text-to-song), and is now working on a Sing-to-Sing technology. But this time, Voicemod decided to explore a different route and not create just another online service and, for the first time, offer its Sing-to-Sing technology into the hands of third-party developers - including music software makers.
As the company explains, "with Voicemod's real-time vocal transformation capabilities, users can now modify their singing voice with unprecedented ease - making music expression accessible to everyone. From adjusting pitch and timbre to applying automatic voice tuning, the technology allows for real-time musical expression, suitable for those wanting to enhance live performances and recording."
The goal is to enable developers to go beyond speech and sound effects by integrating AI-powered singing functionality capabilities. This is done using Voicemod's Software Development Kit (SDK), which already supports Speech-to-Speech, and Soundboard functionalities, allowing third-party developers, including app creators, game developers, music developers, and hardware manufacturers to transform vocal performances in real-time.
The SDK boasts a versatile real-time "AI Voice Tune" suite, offering automatic voice tuning and a variety of effect options like choir, duet, evolver, and calitune. To expand the singing tools, Voicemod also partnered with IP holders, such as Warner and Rovio, to offer exclusive content, providing new revenue streams for authors and rights-holders. The company also says that it was the first in its category to earn Fairly Trained certification for ethically trained AI speech and singing models.
"The technology not only caters to musicians, but also makes musical expression accessible to everyone. It allows users to alter their singing voice instantly, with options to adjust pitch and timbre and even apply automatic voice tuning for flawless performances," the company details.
Voicemod's Sing-to-Sing AI offers potential applications in casual and professional music-making, to interactive concerts and sing-alongs on platforms like Fortnite, Roblox or anywhere online where music and voice play a pivotal role. The tools also address the common hesitation people feel about using their voice online, offering a tool that boosts confidence and fosters inclusion and uninhibited self-expression.
"We're excited to see how creators will incorporate Sing-to-Sing AI technology into their products, whether it's to express themselves, enhance karaoke, or support interactive musical events online; this technology will push their singing to the next level," says Jaime Bosch, CEO and co-founder of Voicemod-
Developers who want to explore the SDK can visit www.voicemod.net/developers/ for more information. Voicemod is headquartered in Valencia, Spain, with offices in the US and Asia.
www.voicemod.net
