Artificial intelligence (AI) audio corporations ElevenLabs said on the 23rd that it added new features to its AI music generation service "ElevenMusic" to help users create music more easily.
The newly added features are "Reference" and "Vocal." Reference generates a new track that matches the mood and style of the AI music generation model "Music v2" when you upload an audio clip 10 seconds to 5 minutes long. It can capture even subtle emotions that were difficult to express with text prompts alone.
Vocal creates a fully original song sung in a given voice by using the user's voice or a library voice. Even people without professional music production experience can easily make a track with just a voice recording, and it can learn a specific voice's tone and singing style to maintain the same vocal across multiple songs.
The "Style" feature, which keeps a song's unique style consistent, now also supports "Music version 2."
An ElevenLabs official said, "All voices and tracks uploaded to the platform are designed to be usable only after passing a strict copyright check before the generation stage," and noted, "Generation that uses another person's copyrighted work is blocked."