Make original music for the things that have to perform: ad beds that sit under a voiceover, product video tracks cut to the edit, podcast themes, jingles with your own lyrics and three-second sonic logos. Start from a track you like and we read it into the prompt first, so you begin from something that already works instead of a blank box.
Switch on vocals, paste your lyrics and use section tags like [Verse] and [Chorus] to shape the song. Lyria sings your words; nothing is borrowed from an existing song.
Music you make here can be used commercially, online and offline: ads, product videos, podcasts, social posts and presentations. Every file is tagged as AI-generated in its metadata, so disclosing AI content on YouTube, TikTok and Instagram is simple. Lyria tracks also carry Google's inaudible SynthID watermark.
A take that fails is refunded. AI Credits never expire and there is no subscription.
Lyria 3.5 for full songs and anything with vocals or your own lyrics. Lyria 3 Clip for quick 30-second pieces. Stable Audio 3 for instrumentals that must be an exact length, such as a bed under a 27-second video.
Yes, in ads, videos, podcasts and social posts, online and offline. It is not cleared for film, TV, radio broadcast or games distribution.
No. Describe the sound instead (era, genre, instruments, mood) or start from a reference track: we read its qualities into the prompt without naming anyone, and the music you make is original.
Yes. Every file is tagged as AI-generated in its metadata, and Lyria tracks also carry Google's inaudible SynthID watermark. Platforms like YouTube ask you to disclose realistic AI content, and this makes that easy.
Yes. Send the track to Add Sound to Video and it is mixed under your clip, ducked under any voiceover and levelled for social.