Voice AI dev weighs regex vs tool-call vs embedding search for a static reply bank with low latency
Snoo_7134 · reddit · 2026-09-14
A developer building a conversational voice AI with a full STT → LLM → TTS pipeline asks how to route fixed questions to a static reply bank without adding latency. He lays out three options with trade-offs:
- A regex matcher to hit the static bank — risks misses and wrong answers.
- Exposing the bank as a tool the LLM can call, with available questions listed in the system prompt; this skips extra hops and TTS but still costs one LLM call — his current favorite.
- Semantic search over the question bank with a locally hosted embedding model, which may add latency depending on hit rate.
He's soliciting other perspectives and real-world experience.
More from coding & agent
- Chrome DevTools tip: block request URLs to test API failure behavior — _jaydeepkarale · 2026-09-14
- Reverse engineering uncovers Anthropic's unreleased Vercel-like PaaS "Antspace" inside Claude Code Web — jedisct1 · 2026-09-14
- AI recreates Camp Nou in 3D with 86,964 individually selectable seats in the browser — shauntrennery · 2026-09-14
- GPT-6 Astra builds a playable 3D samurai scene from one prompt, calling Hyper3D Rodin MCP for assets — JaynitMakwana · 2026-09-14
- Parallel agents stuck queueing on a single test database as scale grows — Clean-Vermicelli-700 · 2026-09-14
- Dev uses Codex as an editor to unify vertical and horizontal video aspect ratios — aziz4ai · 2026-09-14