Adobe just dropped two game-changing AI tools that could solve creators' biggest headaches: finding the right soundtrack and voiceover for videos. The company's new Generate Soundtrack and Generate Speech features in Firefly promise commercially-safe audio generation, putting Adobe ahead of competitors tangled in copyright lawsuits. With creators constantly battling DMCA strikes and expensive licensing fees, Adobe's trained-on-licensed-content approach offers a legal lifeline that could reshape video production workflows.
Adobe just flipped the script on AI-generated audio for creators. The company's rolling out two new tools in its Firefly app that tackle the twin pain points every video creator knows too well: finding the perfect soundtrack and getting quality voiceovers without breaking the bank or running into copyright issues.
Generate Soundtrack hits public beta today, and it's basically like having a music producer who never sleeps. Upload your video clip, and the AI analyzes the footage to suggest thematically appropriate instrumental tracks that sync automatically to your content. You can guide the vibe through preset styles like lofi, hip-hop, classical, or EDM, or just describe what you're after in natural language.
"We wanna help users prompt music. It's a new muscle we need to develop," Alexandru Costin, Adobe's generative AI head, told The Verge. "But we're also offering you this Mad Libs approach where you can pick the vibe, the style, the objective of your clip."
The tool spits out four different variations per prompt, each capped at five minutes. But here's where Adobe's playing chess while others are playing checkers: every track is generated from licensed content, meaning creators won't get hit with copyright strikes that have become the bane of YouTube and TikTok creators everywhere.
"We purchased music and voice from IP owners, that's why we have the confidence to offer it as commercially safe," Costin explained. This licensing-first approach gives Adobe a massive advantage over AI music startups like Suno and Udio, which are currently battling copyright infringement lawsuits after admitting they trained their models on protected materials.
Generate Speech, also launching in public beta, handles the voiceover side with over 50 AI voices powered by Adobe's Firefly Speech Model and ElevenLabs. The tool supports more than 20 languages and lets you fine-tune speed, pitch, and emotion. It even handles pronunciation corrections for names or regional variations - a detail that shows Adobe's thinking about real-world creator workflows.
