NiCE adds agent governance and voice tools to Cognigy
The AMW Read
The release updates an existing agent platform and highlights production controls relevant across enterprise agent deployments.
NiCE adds agent governance and voice tools to Cognigy
NiCE has introduced a new version of Cognigy for building and managing customer service agents. The release adds Agentic Experience Studio, reusable skills, AI-assisted development, decision tracing, evaluation tools, and native speech-to-speech interaction. Teams can connect agents to external tools and systems through MCP and A2A. Existing Cognigy customers can adopt the features while their deployed systems continue to run.
The release places operational control alongside agent creation in the enterprise customer service market. Teams can combine autonomous agents with structured workflows and deterministic logic, choosing where each approach applies. Traces show which skills, tools, and models an agent used, along with token consumption and fallback choices. These features address the gap between building an agent and inspecting its behavior in production, although the announcement does not establish customer outcomes.
For builders evaluating the platform, a concrete test is whether its simulation and comparison tools expose failures before deployment and whether traces provide enough detail to diagnose them afterward. Buyers should also compare model costs: basic AI models are included in conversation pricing, while premium third-party models carry usage-based charges. Customers can retain existing model contracts, and NiCE lists premium options from Deepgram, ElevenLabs, and Anthropic.