ACE Studio

This is the core of the product. You feed it a melody as MIDI plus lyrics as text, and it sings them back with phrasing, vibrato, and breath that follow the tune in a way that sounds less like a machine and more like a take you'd actually keep. Because you control the notes directly, you can rewrite a line, change its timing, or swap a syllable without re-recording anything. Simple as that. The company's own materials describe it as SVS technology, meaning it builds vocals rather than simply swapping an existing voice onto audio.

The voice library covers pop, R&B, jazz, folk, bel canto, kids, and cinematic styles, and the site advertises 100+ voices across 8 languages. As an AI vocal synthesizer, it renders each of those voices on demand rather than shipping fixed samples. Every pre-made voice is royalty-free, so vocals you generate can go into commercial releases, client projects, or content without extra fees or attribution. That licensing model matters if you sell your music. It removes a chunk of clearance work. Sell a track, keep the rights, no paperwork.

Voice cloning lets you turn your own recordings into a custom singing model. You upload clean samples, the system trains on how you sing, and the result sings in your voice across styles. It's built for people who have a sound in their head but can't hit the notes live. Neither can I. Unlike some cloning tools, it uses consented, licensed data. The feature isn't available for plain speech, only singing.

The ACE Bridge plugin drops ACE Studio into your existing workflow as a VST3, AU, or AAX plugin. You trigger vocals and instruments from your DAW over MIDI, and ARA link mode keeps playback and tempo synced. In plain terms, ACE Studio becomes a second window of your DAW instead of a separate app you export files out of. No bouncing files back and forth.

Beyond vocals, the workspace generates instrument parts such as strings and violin. You input MIDI and the AI renders a performance; producers use it to fill out an arrangement when they don't have a real player on hand. It plugs into the same MIDI workflow as the voices, so the learning curve is shared. Same notes, new sound.

Upload an existing track and the Stem Splitter pulls it apart into separate parts, including the vocal. From there, Vocal to MIDI converts that vocal line into editable notes and lyrics. Remixing gets easy. It's the feature that makes remixing and re-singing possible: you can take a finished song and rebuild it, change the key, or layer harmony on top of what's already there without ever tracking a new take.