IHBench: Interruption Handling Benchmark for Voice Agents

IHBench evaluates post-interruption recovery in voice agents executing structured, multi-step workflows. The benchmark contains 45 scenarios to measure if an agent resumes correctly, addresses user interjections, and avoids re-delivering content. It was created by bosonai and last updated on Hugging Face in June 2026.

Use Cases

• Benchmarking voice agent recovery accuracy based on structured workflow scenarios.
• Training models to resume multi-step tasks after user interruptions.
• Evaluating whether agents avoid repeating information already delivered to the user.

Strengths

• The benchmark contains 45 specific scenarios for testing.
• It focuses on a distinct, measurable aspect of agent performance: post-interruption recovery.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Freshness should be verified as the last update was 2026-06-18.

Provenance

Source

bosonai via Hugging Face

Freshness

Last updated 2026-06-18 15:15:10.