Building AI voice and music applications
summary
Teams describe shipping web-first AI voice and music apps on Next.js and Vercel, launching a Grimes-powered app (Elf.tech) that open-sources stems and charges a $9.99 distribution fee for songs that use her name/image, using only company-approved voices for defaults. Work splits roughly 50/50 research and production: fine-tuning open-source models, curator-fed training from music engineers, controlled psychological evaluations, and prototyping controller-style “music gaming” UIs; demos run and are iterated on in the browser, not just promises.
AI-written summary · about
chapters (10)
- 0:00AI voice and music intro
- 1:22Emotion and creativity in AI
- 3:39Voice as future HCI
- 6:11Evaluating AI music/voice
- 8:57Music gaming and UIs
- 11:31Balancing research and production
- 14:18Copyright and ethical data
- 16:16Open vs closed-source AI
- 18:15Future of AI companions
- 21:46Q&A
related talks
34mUsing Appsmith for building GenerativeAI apps
42mMaximising Impact with AI: Ethical Strategies for Developing a Top-Rated Game on a Modest Budget
44mSmartly AI: Cross Platform Voice Application Creation
01h 30mBuilding Your First Multimodal Gen AI App 🚀
23mIntelligence as a Material
46mBecoming an AI engineering company
38mIntegrating AI and Browser API Voice Recognition on Web App
25mConversation with Sean King (SVP, GM at Veritone) and @ianutile