Icon for SLNG

SLNG

Voice agent infrastructure that routes across STT, TTS and LLM providers with region-pinned execution

SLNG is an execution layer for real-time voice agents. It sits between an application and the speech stack, routing across providers (Cartesia, Deepgram, AssemblyAI, Rime, Speechmatics, Resemble, OpenAI and others) behind one API and one set of rules.

Agents are declared as code through UNMUTE, an open standard SLNG publishes for describing prompts, tools, models, language and regions in a single file. The file is framework-agnostic, so the same agent can run on SLNG, Pipecat or LiveKit.

Compliance rules are checked at deploy rather than reported afterwards: a configuration that conflicts with its region fails before going live, and each call is logged with its region, provider and model. SLNG publishes ISO 27001 and SOC 2 Type II certifications alongside HIPAA and GDPR compliance, with a trust centre and regional endpoints across 15 regions.

Pricing: Usage-based

Compliance SOC 2 · HIPAA · GDPR
Screenshot of SLNG webpage

Work on SLNG? Feature it at the top of Audio.

Is your product missing?

Add it here →