Tem Balanco
2.6K posts

Tem Balanco
@tembalanco
Born in 🇿🇦, Proud 🇺🇸, Tech Geek, Husband, Father, Investor, Austinite.














We started this project thinking LMs can’t be prompted to do classification tasks with over 10,000 classes — especially when documents are long! But the incredible @KarelDoostrlnck found this elegant DSPy program that, once optimized on ~50 examples, sets the state of the art.











My next plan for Photo AI is talking people: - add ElevenLabs TTS - add video lipsync That potentially moves it from "fun little app" to "useful B2B product" as companies can use it for: - video ads with talking people - education or explainer videos - talking AI influencers A la HeyGen but 100% AI, which means: - put yourself or your presenter or your influencer in any place/time/event in seconds - let them speak about anything while being there - no need to record training videos like HeyGen, just upload 10 photos No idea if that works but let's try!


MacWhisper 10.0 is available as a free update now! - Supports the new Whisper Turbo model for up to 21x realtime at super high accuracy - Support for local AI models using Ollama - Support for custom OpenAI API endpoints and Groq macwhisper.com







