gptme-voice

Status: active

Voice interface for gptme agents using real-time audio streaming

Updated 2025-09-01
#voice#realtime#audio#gptme#agents#twilio#python#superuser-labs

View on GitHub

Overview

gptme-voice enables real-time voice conversations with gptme agents. It uses the OpenAI Realtime API for low-latency audio streaming, loads agent personality from project config, and dispatches workspace tasks to gptme subagents.

Part of the gptme-contrib monorepo.

Key Features

  • Real-time voice conversations with voice activity detection
  • Agent personality loading from gptme.toml / ABOUT.md
  • Subagent tool dispatches for workspace interaction (read files, check tasks, run commands)
  • Twilio integration for phone call support
  • Local mic/speaker testing client
  • WebSocket-based architecture
  • gptme-contrib - The monorepo containing this package
  • gptme - The framework this integrates with

Categories

  • ai
  • tools