Your music is yours: How RoEx build with AI
Introduction
The emergence of AI means that technology is moving faster than ever before - this is particularly true within the music industry and creator tools sector.
We understand that music artists and producers often feel confused and wary of the technology in front of them, and the companies that are building it.
Actions by certain bad actors within our industry - companies that have used artists' music to train AI models without permission - mean that musicians need to be able to trust the products they are using.
Musicians need to know that their music and creativity will be kept safe when creating.
With that in mind we hope that this overview gives you a clear understanding of how RoEx use AI and what this means to you as an artist.
Our Approach
RoEx was created with the belief that better sound should be available for all artists.
We believe that AI-powered tools should support the existing creative process for music creation, rather than replace it completely.
Helping you improve is central to everything we build - with educational features representing core elements within our tools. We want artists to develop their production skills, rather than handing everything over to technology.
As we grow, we wish to keep lowering the bar for better sound for DIY artists, but also give professional artists and producers time-saving tools - allowing full focus to be kept on the creative input for the work.
Transparency
When using RoEx products your music will always remain yours.
We do not train our AI models using any audio that is uploaded to any of our products or tools.
Our AI models are trained in-house, by engineers at RoEx.
There are no generative AI tools within RoEx products.
We are proud to be a signatory to the Roland AI Initiative - AI for Music - a set of principles developed for responsible AI use in music.
Responsibility
Running AI models at scale uses energy, and we take that seriously.
RoEx's AI models are purpose-built for specific audio processing tasks, unlike large language models, which require significantly more energy to run.
But even so, we are not going to pretend that our models have no environmental footprint - they do.
To try and counter this we choose data centre providers with strong environmental commitments wherever we can, and contribute to climate action through Stripe Climate - where a portion of all RoEx revenue goes directly to carbon removal projects.
The Future
As RoEx continues to grow and continues to build, our aim is to get closer to the creative workflow, and place our tools at the heart of the creative work we care about.
As we do so we will continue to hold ourselves accountable to the above principles and practices.
Please reach out to us if you have any questions for us: info@roexaudio.com.
Thank you,
Dave and the RoEx team