Benjamin Genchel

Richard Savery, Benjamin Genchel, Jason Smith et al.

Harriet Padberg wrote Computer-Composed Canon and Free Fugue as part of her 1964 dissertation in Mathematics and Music at Saint Louis University. This program is one of the earliest examples of text-to-music software and algorithmic composition, which are areas of great interest in the present-day field of music technology. This paper aims to analyze the technological innovation, aesthetic design process, and impact of Harriet Padberg's original 1964 thesis as well as the design of a modern recreation and utilization, in order to gain insight to the nature of revisiting older works. Here, we present our open source recreation of Padberg's program with a modern interface and, through its use as an artistic tool by three composers, show how historical works can be effectively used for new creative purposes in contemporary contexts. Not Even One by Molly Jones draws on the historical and social significance of Harriet Padberg through using her program in a piece about the lack of representation of women judges in composition competitions. Brevity by Anna Savery utilizes the original software design as a composition tool, and The Padberg Piano by Anthony Caulkins uses the melodic generation of the original to create a software instrument.

Justin Zhao, Flor Miriam Plaza-del-Arco, Benjamin Genchel et al.

As Large Language Models (LLMs) continue to evolve, evaluating them remains a persistent challenge. Many recent evaluations use LLMs as judges to score outputs from other LLMs, often relying on a single large model like GPT-4o. However, using a single LLM judge is prone to intra-model bias, and many tasks - such as those related to emotional intelligence, creative writing, and persuasiveness - may be too subjective for a single model to judge fairly. We introduce the Language Model Council (LMC), where a group of LLMs collaborate to create tests, respond to them, and evaluate each other's responses to produce a ranking in a democratic fashion. Unlike previous approaches that focus on reducing cost or bias by using a panel of smaller models, our work examines the benefits and nuances of a fully inclusive LLM evaluation system. In a detailed case study on emotional intelligence, we deploy a council of 20 recent LLMs to rank each other on open-ended responses to interpersonal conflicts. Our results show that the LMC produces rankings that are more separable and more robust, and through a user study, we show that they are more consistent with human evaluations than any individual LLM judge. Using all LLMs for judging can be costly, however, so we use Monte Carlo simulations and hand-curated sub-councils to study hypothetical council compositions and discuss the value of the incremental LLM judge.

Yin-Cheng Yeh, Wen-Yi Hsiao, Satoru Fukayama et al.

Several prior works have proposed various methods for the task of automatic melody harmonization, in which a model aims to generate a sequence of chords to serve as the harmonic accompaniment of a given multiple-bar melody sequence. In this paper, we present a comparative study evaluating and comparing the performance of a set of canonical approaches to this task, including a template matching based model, a hidden Markov based model, a genetic algorithm based model, and two deep learning based models. The evaluation is conducted on a dataset of 9,226 melody/chord pairs we newly collect for this study, considering up to 48 triad chords, using a standardized training/test split. We report the result of an objective evaluation using six different metrics and a subjective study with 202 participants.

Benjamin Genchel, Ashis Pati, Alexander Lerch

Deep generative models for symbolic music are typically designed to model temporal dependencies in music so as to predict the next musical event given previous events. In many cases, such models are expected to learn abstract concepts such as harmony, meter, and rhythm from raw musical data without any additional information. In this study, we investigate the effects of explicitly conditioning deep generative models with musically relevant information. Specifically, we study the effects of four different conditioning inputs on the performance of a recurrent monophonic melody generation model. Several combinations of these conditioning inputs are used to train different model variants which are then evaluated using three objective evaluation paradigms across two genres of music. The results indicate musically relevant conditioning significantly improves learning and performance, and reveal how this information affects learning of musical features related to pitch and rhythm. An informal subjective evaluation suggests a corresponding improvement in the aesthetic quality of generations.