cekura/s2s-agent-bench

Cekura Speech-to-Speech Agent Benchmark 82 phone-call test cases for voice agents that work through tools: 59 at a clinic appointments desk and 23 on a Medicare sales line. Each case is a simulated caller with a goal, a script that branches on what the agent does, and the result the call should reach: the outcome in words and the tool calls the agent should make. This is the test set behind the speech-to-speech board at benchmarks.cekura.ai/speech-to-speech. The harness and the… See the full description on the dataset page: https://huggingface.co/datasets/cekura/s2s-agent-bench.

<iframe src="https://silytics.si/embed/item/58381f0f-3d90-4001-8035-75c70a978da6" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.