Cekura

Cekura is a web-based automated QA and observability platform for enterprises building and deploying voice AI and chat AI agents.

Key capabilities include:

• Pre-production call and conversation simulations that test end-to-end voice pipelines, including speech input, instruction-following, tool calls, and multi-turn task completion across diverse personas.
• Automated scenario generation and replay using scripted cases, AI-generated scenarios, and production conversation replays to cover core flows and known failure points.
• Edge-case testing for voice agents, including accents, interruptions, silence, and background noise to validate real-world conversational quality.
• Production monitoring and analytics with dashboards and alerts, tracking metrics such as latency, task success, WER, interruption handling, and tool-call metadata.

Pricing: Freemium with a 7-day free trial period (300 credits). Paid plans include Developer ($30/month with 740 credits and then pay-as-you-go), and Enterprise (custom pricing).

Notable for combining pre-release evaluation and production observability for conversational agents in one workflow, with support for regression and A/B testing in CI/CD pipelines. Formerly known as Vocera.