提示词与规则：语音用户模拟器中语音自然度行为的审计与控制
Prompts versus Rules: Auditing and Controlling Speech Naturalness Behaviors in Voice User Simulators
浏览论文内容
中文总结 AI 辅助
本研究对比了基于LLM的提示词方法与基于规则的算法在语音用户模拟器中生成语音自然度行为的效果，发现规则算法更优，提出需语言学导向方法缩小真实用户行为模拟的差距。
中文摘要 AI 辅助
随着语音智能体在商业领域日益普及，用于评估已部署智能体的用户模拟器也在不断发展，以纳入更贴近现实、更多样化的语音自然度行为，包括不流畅表达、打断和反馈回应。用户模拟器的质量直接影响智能体评估结果的有效性，但我们发现，迄今为止大多数研究并未详细考察这些行为的预期配置是否在模拟中得以实现。本研究中，我们针对tau-Voice（我们自身基于大语言模型（LLM）的提示方法，涉及三个模型）以及我们针对不流畅表达、打断和反馈回应的基于规则的注入算法，对其已实现的自然度行为进行审计。我们发现，通过提示词实现这些行为的方式不可靠，产生的语音与指令不一致，其位置和分布的自然度低于指令所暗示的水平。相比之下，我们的基于规则、与模型无关的算法能够生成可控且多样的自然度行为，更符合自然语音的特征。我们的结果表明，未专门针对用户模拟进行训练的LLM本身不足以代表真实的用户行为，需要采用语言学导向的确定性方法或专用模型来缩小这一差距；审计并报告已实现的自然度行为而非配置设置，是使这一差距显现的关键。
英文摘要
As voice agents gain more popularity commercially, the user simulators used to evaluate the deployed agents are also being developed to include more realistic, variable, and diverse speech naturalness behaviors -- disfluency, interruption and backchanneling. The quality of the user simulator directly affects the validity of agent evaluation results. However, we find that most studies so far have not examined in detail whether the intended configuration for these behaviors is realized in the simulation. In this study, we audit the realized naturalness behaviors of tau-Voice, our own LLM-based prompting approach across three models, and our rule-based injection algorithm for disfluency, interruption, and backchanneling. We find that prompting for these behaviors is unreliable and produces speech inconsistent with the instructions, placed and distributed less naturally than the instruction implies. In contrast, our rule-based, model-free algorithm produces controllable and diverse naturalness behaviors more aligned with natural speech. Our results suggest that LLMs not purpose-trained for user simulation are not sufficient on their own to represent authentic user behavior, and that linguistically informed deterministic approaches or specialized models are needed to close the gap; auditing and reporting realized naturalness behaviors, rather than configured settings, is what makes that gap visible.
发表机构
- Dialpad Inc.（戴尔德公司）
机构由 AI 辅助整理，请以论文原文为准。
