The notion of a musical repertoire's "sound" is frequently evoked in journalism and scholarship, but what parameters comprise such a sound? This question is addressed through a statistically-driven corpus analysis of hip-hop music released during the genre's Golden Age era. The first part of the paper presents a methodology for developing, transcribing, and analyzing a corpus of 100 hip-hop tracks released during the Golden Age. Eight categories of aurally salient musical and production parameters are analyzed: tempo, orchestration and texture, harmony, form, vocal and lyric profiles, global and local production effects, vocal doubling and backing, and loudness and compression. The second part of the paper organizes the analysis data into three trend categories: trends of change (parameters that change over time), trends of prevalence (parameters that remain generally constant across the corpus), and trends of similarity (parameters that are similar from song to song). These trends form a generalized model of the Golden Age hip-hop sound which considers both global (the whole corpus) and local (unique songs within the corpus) contexts. By operationalizing "sound" as the sum of musical and production parameters, aspects of popular music that are resistant to traditional music-analytical methods can be considered.

中文翻译：

---

寻找黄金时代的嘻哈音乐（1986–1996）

音乐剧“声音”的概念在新闻学和学术界经常被提及，但是，什么参数构成了这种声音呢？通过对该类型的黄金时代时代发布的嘻哈音乐进行统计驱动的语料库分析，可以解决此问题。本文的第一部分介绍了一种方法，用于开发，转录和分析在黄金时代发行的100条嘻哈曲目的语料库。分析了八种主要的听觉音乐和生产参数：速度，编排和纹理，和声，形式，声音和歌词配置文件，全局和局部生产效果，声音加倍和支持以及响度和压缩。本文的第二部分将分析数据分为三个趋势类别：变化趋势（随时间变化的参数），流行趋势（整个语料库中通常保持不变的参数）和相似性趋势（每首歌之间相似的参数）。这些趋势形成了黄金时代嘻哈声音的通用模型，该模型同时考虑了全局（整个语料库）和本地（语料库中的唯一歌曲）上下文。通过将“声音”作为音乐和生产参数的总和进行运算，可以考虑对传统音乐分析方法有抵抗力的流行音乐的各个方面。