RVCBench benchmark evaluates robustness of 18 voice cloning models; accepted at NeurIPS 2026
Researchers from the University of British Columbia and the Vector Institute introduced RVCBench, a benchmark that systematically evaluates the robustness of 18 open-source voice cloning models across 18 evaluation metrics and 10 task types, covering input, generation, output, and audio perturbation stages. The benchmark is packaged as a pip-installable toolkit and has been accepted at NeurIPS 2026.
Coverage timeline
机器之心机器之心
导读：只需几秒参考音频，现代语音克隆模型就能合成与目标说话人高度相似的语音。但真实部署中的参考录音往往带有噪声，文本可能异常、超长或跨语言，生成的语音还要经过压缩和电话传输，有人甚至会给自己的声音加上「防克隆」保护。在这些条件下，模型还能保持内容、身份和自然度吗？来自英属哥伦比亚大学（UBC）与 Vector Institute 的研究团队提出 RVCBench，把语音克隆的鲁棒性拆成输入、生成、输出和音频扰动四个环节，用 18 项评测、10 类任务系统检验 18 个代表性开源模型，并将基准升级为可直接 pip 安装的通用语音克隆评测工具包。该工作已被 NeurIPS 2026 接收。 论文标题：RVCBench: Benchmarking the Robustness of Voice Cloning Across Modern Audio Generation Models 会议：NeurIPS 2026 论文链接：https://arxiv.org/abs/2602.00443 代码链接：https://github.com/Nanboy-Ronan/RVCBench（PyPI：rvcbench） 项目主页：https://nanboy-ronan.github.io/RVCBench/ 背景：当语音克隆走出实验室 语音克隆，也被称为零样本语音合成（zero-shot TTS）或说话人条件语音合成，目标是只凭几秒参考音频，就生成与目标说话人音色相近的任意内容语音。过去两年，以自回归编解码 token 语言模型和扩散 / 流匹配模型为代表的新一代方法快速发展，在干净测试集上的相似度和自然度已经相当惊人，个性化语音助手、影视配音等应用也随之落地。 但「在测试集上表现好」与「在真实场景中可靠」是两回事。真实部署中，模型会不可避免地遇到这样一些情况： 参考录音来自嘈杂环境，或混入了其他说话人的声音； 输入文本带有异常格式、幻觉式内容，甚至是诈骗类话术； 需要生成很长的段落，或跨语言克隆； 生成的语音还要经过 MP3 / AAC / Opus 压缩，或通过窄带电话传输； 声音的主人主动给录音加上对抗扰动，试图让自己的声音「无法被克隆」。 现有评测大多聚焦干净条件下的生成质量，对上述部署偏移下的鲁棒性缺少系统、统一的刻画。RVCBench 想回答的问题，正是从「生成质量有
Related stories
WorldMemArena benchmark evaluates multimodal agent memory via action-world interaction
Researchers from UCSB, Stanford, ETH Zurich, Johns Hopkins, CMU, and J.P. Morgan Chase released WorldMemArena, a multimodal agent memory benchmark accepted at NeurIPS 2026. It evaluates memory through an observation-action-environment feedback loop, breaking long-term memory into write, maintain, retrieve, and use stages to pinpoint where agents fail. The benchmark includes multimodal cues, memory lifecycle tracking, and world changes after actions, addressing limitations of static recall or final QA evaluations.
1 reports · 1 sources
Fudan, Tencent Hunyuan, Tsinghua Release ExplorationBench to Test AI Rule Discovery
Fudan University, Tencent Hunyuan, and Tsinghua University released ExplorationBench, a benchmark measuring how AI systems discover rules in unfamiliar environments. It includes two altered 'alien worlds'—the programming language AlienCode and the logic proof system AlienLogic—where models must learn rules through trial and error. Results showed that without environment feedback, scores remained low (max 11.0%), while autonomous exploration led to high scores (up to 89.0%).
Anthropic: Small AI models may self-replicate on IoT devices within months
Anthropic researchers report that small AI models may be months away from self-replicating on commercial IoT devices like smart ovens and off-brand Android TV boxes. In tests, a 9B model ran inference on 8 of 9 hardware types, and GLM-5.2 autonomously spread an 8B model to all 8 nodes in 4 of 5 attempts. The researchers predict models under 30B could achieve end-to-end replication by the end of 2026.
UCLA and Adobe open-source Tev-27B long-horizon agent model, matching GPT-6 Astra and Claude Opus 5 on ALE-CM
Researchers from UCLA and Adobe Research released a paper and open-sourced Tev-27B, a long-horizon agent model trained on their new Tev-Gym benchmark with dense, verifiable, progressive rewards via progress potential reward shaping (PPRS). On the Agents' Last Exam computation and mathematics subset (ALE-CM), Tev-27B achieves Pass@5 of 0.28, matching GPT-6 Astra and Claude Opus 5; the same recipe applied to Tev-320B makes it the strongest open-source model on Terminal-Bench 4.0's computation and mathematics subset.
News updates every hour. Subscribe in your RSS reader to get new stories automatically.
Subscribe