GoKawiilHacker News user asks why audio-plus-text to audio AI tools remain scarce
A poster on Hacker News describes searching for a commercial AI model that accepts both a reference sound and a text prompt to generate new audio, similar to how image models use reference images plus prompts. They report trying ElevenLabs' text-to-SFX tool and Stable Audio 3, saying neither handled reference audio well or produced accurate results for their sound-effect generation use case.