AI Hits Different - A Music Album From the Machine Room
summary
Claims a full album is produced in weeks using a concrete pipeline: large language models generate lyric drafts with a personal corpus, Suno AI → Splitter AI → Basic Pitch pipeline turns generated audio into stems and MIDI, MidJourney and Sora supply visuals and art, and AI mastering engines finish tracks. The process averages 40–50 generations per final song, halves total production time (from ~80 to ~40 hours), yields playable club tracks, and triggers platform moderation (Spotify flag).
AI-written summary · about
chapters (17)
- 0:00AI and music project intro
- 6:46AI in music motivation
- 10:24Suno AI music generators
- 21:16AI audio quality challenges
- 24:27AI-assisted lyric writing
- 27:23Generating melodies refining
- 30:17AI song to human production
- 34:07Human creativity in mixing
- 37:25AI mixing and mastering tools
- 39:12Evo on AI in music
- 46:38AI for demos and inspiration
- 49:24Ethics and AI music future
- 50:27Authenticity and live performances
- 53:15Time savings with AI tools
- 57:16Spotify AI moderation issues
- 59:04AI creates options humans choose
- 1:01:12Q&A
related talks
55mAI Music Generation, From Prompt to Production
14mAI & Music: New Opportunities for Creators
52mAI Lofi HipHop - How (Not) To Generate AI Music
36mWhen your next hit is co-composed by an AI
44mCoding to the Beat: When AI Joins the Band
44mWith AI, Does Music Have the Blues?
30mDeveloping AI that enhances human creativity
26mInfinite AI