Voice cloning company ElevenLabs has unveiled a groundbreaking tool that enables users to generate sound effects using simple prompts. First announced in February, this innovative tool is now accessible to all users.
Starting today, users can enter prompts such as “waves crashing,” “metal clanging,” “birds chirping,” and “racing car engine” to create short sound snippets. Additionally, the tool can produce instrumental musical clips up to 22 seconds long with prompts like guitar loops, jazz saxophone solos, and techno music loops.
Free-tier users have a monthly allowance of 10,000 characters for generating sounds, with each sound request consuming approximately 150 characters. This translates to about 60 sound effects per month. However, free-tier users must credit “elevenlabs.io” when publishing any content featuring these sound clips.
ElevenLabs revealed that its model was trained using Shutterstock’s extensive audio library of licensed tracks. The tool, tested during its alpha phase by video game developers, film producers, social media content creators, and marketers, has already demonstrated significant potential.
The tool enforces strict guidelines against generating sounds that breach its Prohibited Content and Uses Policy, covering areas such as self-harm, threats to child safety, and fraudulent activity.
Although relatively few companies focus on AI-driven sound generation, the music creation sector is more competitive. Notable players include Stability AI-backed Harmonai with Dance Diffusion, Google’s MusicLM, OpenAI’s Jukebox, and Meta’s AudioCraft model. Additionally, TikTok and Adobe have explored generative AI-based music creation tools.
By offering this new sound effects generator, ElevenLabs positions itself as a key player in the evolving landscape of AI-powered audio technology.
