Voice-Swap is today introducing Stem-Swap – a powerful new tool that lets you upload fully mixed tracks and transform the vocals into the voices of our licensed singers. Stem-Swap automatically splits the stems of your uploaded file, and allows you to change the voice, adjust the levels, add FX and more – you can even download the stems as a single mixed track or the individual stems, with your new vocals intact.
Stem-Swap is designed for artists, producers, songwriters and vocalists to experiment with new voices in their tracks, without having to export acapellas or re-import new voices back into their DAW. Using Step-Swap, music-makers can demo new voices to figure out which one they’d like to use, quickly and easily, directly in the browser. Subscribers can then download their new idea.
You can even ‘Split’ the vocal tracks, and hear the original vocal and the new AI version at the same time, mixing the levels and adding FX to each individually. Want to hear what Angie Brown or Robert Owens might sound like singing backing vocals on your new track? Now you can, with only one upload required.
As with everything Voice-Swap, our artists are 100% licensed by us, and we work with them and their teams to record all the training data that goes into their models. When you download either the stems or the fully mixed track from Stem-Swap, our artists are paid a 50/50 royalty split, ensuring that real talent is compensated at every step.
We believe that creative and ethical AI can co-exist and that these ground-breaking tools can be made available to all, without exploiting the artists whose talent goes into building the models.
Stem-Swap is a perfect example of revolutionary tech and creative potential, paired with traditional and equitable royalty models. We believe the future for AI and artistry can be bright if those building the tools respect artists, rightsholders and IP.