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.
By Anupam Purwar, Shashank Singh, Kritika Srivastava
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.
By Aryan Vijay Bhosale, Harshit Rajgarhia, Akhil Pothanapalli, Asif Shaik, Abhishek Mukherji, Dinesh Manocha
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.
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.
By David Ayllon, Alice Baird, Jeffrey Brooks, Franc Camps-Febrer, Jakub Piotr C{\l}apa, Theo Lebryk, Jens Madsen, Olya Ossipova, Sharath Rao, Hoon Shin, Tigran Soghbatyan, Georg Streich, Rashish Tandon, Panagiotis Tzirakis
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.
By Jiajun Fan, Jingyuan Li, Prashanth Gurunath Shivakumar, Jia-Hong Huang, Qi Luo, M. Maruf, Ivan Bulyko, Ge Liu, Roger Ren
Today, we’re releasing new tools to help developers go from prototype to production faster: AgentKit, expanded evals capabilities, and reinforcement fine-tuning for agents.
arXiv:2606. 15141v1 Announce Type: cross Abstract: While LALMs show promise on audio question answering, they fail to focus on question-relevant segments of audio and provide a clear, checkable reasoning process when dealing with complex audio reasoning.
By Siyuan Zhang, Jian Zong, Junyu Wang, Peiyuan Jiang, Jiahao Yan, Jingyu Zhang, Tianrui Wang, Xiaobao Wang, Longbiao Wang, Jianwu Dang
arXiv:2510. 25577v2 Announce Type: replace-cross Abstract: Recent advances in Speech Foundation Models (SFMs) enable direct processing of raw audio, allowing models to respond to subtle paralinguistic variation.
By Harm Lameris, Shree Harsha Bokkahalli Satish, Joakim Gustafson, \'Eva Sz\'ekely
arXiv:2607. 06713v1 Announce Type: cross Abstract: Large language models are rapidly moving towards closing the development cycle, transitioning from simple assistive companions to autonomous contributors deeply embedded into collaborative development environments.
By Razvan Mihai Popescu
We’re sharing lessons from a small scale preview of Voice Engine, a model for creating custom voices.