Show HN: Fusion-runtime – self-hosted voice agents, STT+LLM+TTS in one process

Category: ai-ml

Tags: voice-agent, speech-to-text, text-to-speech, llm, real-time, self-hosted

Score: 7.5/10 (Innovation: 7, Technical: 8, Documentation: 8, Utility: 7)

Fusion-runtime is a self-hosted voice agent runtime that integrates speech-to-text, LLM, and text-to-speech into a single process with low-latency streaming. It provides a simple Python API, tool calling, browser integration, and detailed performance benchmarks, making it a practical solution for building real-time voice agents.

Target audience: AI engineers, backend developers, and Machine Learning practitioners building voice interfaces

Repository: https://github.com/SamarthUrs18/fusion-runtime · Python · Apache-2.0 · 4 stars

View on Hacker News