Welcome Mike, glad you’re here. That’s actually a really useful path to document: Agent Zero, Hermes, OpenHuman, local models, then speech-to-text and a second brain setup. A lot of people get stuck exactly where you did, so sharing what worked and what didn’t will help the whole group. For speech-to-text, I’d probably start simple before going deep: Whisper-based tools are usually the most reliable first experiment. And on OpenBrain vs LLM Wiki vs Second Brain, the best choice might depend on your goal: personal knowledge base, AI agent memory, or team-style documentation. Curious what you want it to do first?