Large Language Model (LLM) judges exhibit strong reasoning capabilities but are limited to textual content. This leaves current automatic Speech-to-Speech (S2S) evaluation methods reliant on opaque and expensive Audio Language Models (ALMs). In this work, we propose TRACE (Textual Reasoning over Audio Cues for Evaluation), a novel framework that enables LLM judges to reason over audio cues to achieve cost-efficient and human-aligned S2S evaluation. To demonstrate the strength of the framework, we first introduce a Human Chain-of-Thought (HCoT) annotation protocol to improve the diagnostic capability of existing judge benchmarks by separating evaluation into explicit dimensions: content (C), voice quality (VQ), and paralinguistics (P). Using this data, TRACE constructs a textual blueprint of inexpensive audio signals and prompts an LLM to render dimension-wise judgments, fusing them into an overall rating via a deterministic policy. TRACE achieves higher agreement with human raters than ALMs and transcript-only LLM judges while being significantly more cost-effective. We will release the HCoT annotations and the TRACE framework to enable scalable and human-aligned S2S evaluation.

翻译：大型语言模型（LLM）评判者展现出强大的推理能力，但仅限于处理文本内容。这导致当前自动语音到语音（S2S）评估方法依赖于不透明且昂贵的音频语言模型（ALM）。本研究中，我们提出了TRACE（基于音频线索的文本推理评估），这是一个新颖的框架，使LLM评判者能够基于音频线索进行推理，从而实现经济高效且与人类对齐的S2S评估。为展示该框架的优势，我们首先引入了一种人类思维链（HCoT）标注协议，通过将评估分解为明确维度——内容（C）、语音质量（VQ）和副语言学特征（P），以提升现有评判基准的诊断能力。利用这些数据，TRACE构建了低成本音频信号的文本蓝图，并提示LLM进行维度级评判，再通过确定性策略将其融合为总体评分。TRACE与人类评分者的一致性高于ALM及仅基于转译文本的LLM评判者，同时成本效益显著更高。我们将发布HCoT标注数据与TRACE框架，以支持可扩展且与人类对齐的S2S评估。