Storytelling keeps evolving, and sound is a powerful part of that journey. With Generate Sound Effects (GenSFX) in Adobe Firefly, creators can describe any sound—or perform it with their own voice—and instantly generate results that sync perfectly to video. Behind this new feature are Adobe researchers who set out to transform one of the most time-consuming parts of audio work into something more accessible (and fun) for everyone.
GenSFX gives everyday creators and professional sound designers alike a faster, more intuitive way to enhance their videos with the sounds they imagine. The most innovative aspect is voice performance, a feature that allows creators to guide the AI through their own vocal expressions. The tool captures the spontaneity of human creativity while generating results that are both usable and unique. Instead of standardizing audio, GenSFX introduces variety and personalization into sound design, empowering creators to move beyond stock libraries and add their own voice, literally, to their work.
A close collaboration between Adobe Research and product teams, GenSFX evolved from an experimental concept into a fully realized tool for creators everywhere in just one year (the project first debuted as Project SuperSonic at MAX Sneaks in October 2024). The journey reflects how innovation at Adobe often takes place: by bridging research insights with product design to create tools that put more creative power in people’s hands.
Watch the full video to hear from researchers Justin Salamon and Oriol Nieto, along with staff designer Adolfo Hernandez Santisteban, as they share why they took on this challenge and how their vision is shaping the future of sound design. Then, experience the magic and bring your imagination to life— make your own sound effects with GenSFX at https://firefly.adobe.com/generate/sound-effects !
Want to explore more innovation from Adobe Research? Discover our latest news here.