提出新指标评估伴奏生成是否贴合输入音频，更准更可靠。
Accompaniment Prompt Adherence: A Measure for Evaluating Music Accompaniment Systems
- 基于分布匹配设计新评估方法，衡量伴奏与提示音的契合度。
- 实验显示该指标与人听判断高度一致，能有效区分质量下降情况。
- 开源实现支持快速对比不同生成模型性能，适合研究者使用。
音乐伴奏生成系统快速发展，但缺乏标准化指标来评估生成结果与条件音频提示的对齐程度。本文提出一种基于分布的度量方法——伴奏提示一致性（Accompaniment Prompt Adherence, APA），并通过合成数据扰动的客观实验和人类听觉测试进行验证。结果表明，APA 与人类对一致性的判断高度吻合，且对降低一致性的变换具有良好的区分能力。我们基于广泛使用的预训练 CLAP 嵌入模型发布了 APA 的 Python 实现，为评估和比较伴奏生成系统提供了一个有价值的工具。
原文摘要 · Abstract (English)
Generative systems of musical accompaniments are rapidly growing, yet there are no standardized metrics to evaluate how well generations align with the conditional audio prompt. We introduce a distribution-based measure called "Accompaniment Prompt Adherence" (APA), and validate it through objective experiments on synthetic data perturbations, and human listening tests. Results show that APA aligns well with human judgments of adherence and is discriminative to transformations that degrade adherence. We release a Python implementation of the metric using the widely adopted pre-trained CLAP embedding model, offering a valuable tool for evaluating and comparing accompaniment generation systems.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。