ScalafromHackernoon3 months agoThe Minimalist's Guide to Speech-to-Text: Big Wins with Little Data | HackerNoonUtilizing a text-only LLM for speech-to-text DE significantly enhances performance while using less training data.