Realtime speech-to-speech bridge for Syrinx — OpenAI Realtime, Gemini Live, and openai-compatible adapters; the Responder-Thinker delegate seam

MemberJunction: RingCentral Realtime Bridge driver (telephony). Connects the realtime agent engine to a phone call over a registered SIP softphone (ringcentral-softphone / SIP-TLS + RTP-SRTP) — the only RingCentral transport carrying bidirectional call au

Continuous voice for Hermes Agent—keep talking while durable tasks run in the background.

Real-time voice conversations for the Claude CLI — talk to Claude naturally from your terminal.

Sipgate AI Flow integration for Jovo Framework

Platform-agnostic core for OtterVoice: VoiceSession, state machine, provider router, events, and mock providers.

EpisCloud coding agent SDK — typed tools, permissions, streaming, sub-agents, handoffs, guardrails, structured outputs.

Give your Claude Code a phone number. Talk to it, and it calls you back when the work's done.

React Native / Expo audio backends for the Glove voice pipeline — on-device mic capture, PCM playback, and Silero VAD

Vercel AI SDK bridge for Syrinx — drive any ai@6 model or agent as the voice pipeline's reasoner, with opt-in speculative generation

Agent-facing CLI for Syrinx — run a turn or a text exchange with deterministic, machine-readable (--json) output and exit codes. Not a console: no REPL, no chat loop, no microphone.

Node backend + signalling server for @llmrtc/LLMRTC

MCP server that gives Claude Code a conversational voice — speaks short, peer-style summaries aloud, ducking your music so it's audible. Two engines (Windows SAPI + ElevenLabs), works from WSL and native Windows.

Live & generative rich-media for @warlock.js/ai — ai.realtime() duplex voice sessions and ai.video() text-to-video. The heavy, demand-gated modalities, kept out of core so text/image/speech stay dependency-light.

Official TypeScript/JavaScript SDK for PyAI, speech-to-text (Hear), text-to-speech (Speak), realtime voice agents (Omni), and call compliance (Trace).

Node-RED nodes for jambonz platform

An MCP server to allow LLMs to speak and listen via bidirectional voice loops

Discord voice bridge for CLI coding agents.

Built-in spec templates per target shape (Section 26 Studio)

CallMCP BYOK driver: Twilio transport (calls/numbers/SMS) bridged to a bring-your-own-key realtime brain (OpenAI Realtime API today, Grok Voice Agent API wire-compatible adapter stubbed). driver_id: twilio_openai. Normative reference: SPEC.md at the repo