Optimizing Voice AI Latency with Self-Hosted Models
How we reduced time-to-first-audio from 5 seconds to 1 second using sentence-level streaming with Ollama, Whisper, and ElevenLabs on self-hosted infrastructure.
Tag: voice-ai
We Made AutoMem Work With Voice AI (And It’s Pretty Cool)
We built an SSE sidecar that lets AutoMem work with ChatGPT, Claude mobile, and ElevenLabs voice agents. 322 lines of Node.js. Voice AI with persistent memory. Running on Railway for $5/month. It’s actually pretty cool.