A New Framework for Evaluating Voice Agents (EVA)

Benchmarking LLM Judges for Voice-Agent Evaluation: Reliability, Calibration, and Human Oversight

The paper evaluates the use of large language models (LLMs) as judges for assessing conversational voice agents, comparing human judgments with GPT‑4.1 and GPT‑5 across telecom and retail interactions. It examines agreement, metric‑level correlations, and consistency across three evaluation configurations (p0, p1, p2) to determine how reliably LLMs can judge conversational quality and safety. The study finds that LLM‑based evaluation can be effective but its reliability varies by metric and configuration, suggesting a hybrid approach where LLMs handle scalable assessment while humans focus on metrics requiring contextual interpretation.

DuplexWorld: Can voice agents help you get through the day?

arXiv:2608. 10716v1 Announce Type: cross Abstract: Speech-to-speech (S2S) voice agents are increasingly being incorporated into enterprise for customer care and as daily companions for consumers owing to the ease of the conversational modality over text.

Introducing Real World VoiceEQ: Measuring the human quality of voice AI

Introducing next-generation audio models in the API

For the first time, developers can also instruct the text-to-speech model to speak in a specific way—for example, “talk like a sympathetic customer service agent”—unlocking a new level of customization for voice agents.

RW-Voice-EQ Bench: A Real World Benchmark for Evaluating Voice AI Systems

arXiv:2607. 14846v1 Announce Type: cross Abstract: Current voice AI benchmarks typically evaluate isolated capabilities such as speech intelligibility, word error rate, or text-based dialogue quality, but they rarely test whether systems harness the acoustic information that distinguishes spoken language from its textual representation.

SpeechGym: An Audio-Native Gym for Training Voice Agents via Reinforcement Learning

SpeechGym is an audio‑native environment that lets two omni‑modal models converse entirely in native audio, eliminating external ASR/TTS and API boundaries while preserving the tasks, tools, and success checks of a standard text‑based agent benchmark. By training end‑to‑end, the framework addresses perceptual failures—such as misheard arguments that cascade into failed calls—and behavioural failures, both of which are automatically labeled for free. Using per‑turn process rewards to overcome reward sparsity, agents trained in SpeechGym transfer to an independent voice benchmark, doubling task success and improving efficiency in turns and tokens.