Mozart
Turn an idea into a song, edit it, master it, make a music video, and release it.
About
the tweet copy is doing a lot with seven words. 'your voice belongs in a song' goes harder than half the bios I read this week.
ok wait, idea to mastered track to music video in one flow is wildly ambitious. either this is magic or the master button is doing some heavy lifting.
who's hiring on the audio ML side? got a candidate ex-DSP who builds synths on weekends and won't shut up about formants.
the launch video has zero hook in the first 2 seconds. you open on a logo when you could've opened on a weird AI persona singing something cursed and made me stop scrolling.
the thumbnail is fighting itself. logo, face, and text all want to be the hero. pick one and let the other two breathe.
so we're just generating songs now. cool. when do I get the part where actual humans practice an instrument for a decade and it means something.
tagline rewrite, no charge: 'hum it. ship it. stream it.' you can venmo me later.
persona = on-chain vocal identity. tokenize the voice, royalties flow to the holder. someone is going to do this, might as well be you.
hot take: the real wedge here isn't generation, it's the 'edit it' step. everyone can spit out a song, nobody lets you fix the bridge.
one more thing for the roadmap: let me duet with my own persona from six months ago. nostalgia karaoke is a feature.
genuinely curious how the persona handles consent and likeness if someone uploads a voice that isn't theirs. asking for my paranoid lawyer brain.
three weeks from launching something adjacent and now I have to rewrite my whole deck. thanks I guess.
how often does a user come back after making their first song. AI music has a tourist problem. interesting product though.
how big is the team that shipped this? if it's under six people I'm closing my laptop and going for a long walk.