Adaptive Background Music for a Fighting Game: A Multi-Instrument Volume Modulation Approach

TL;DR

This work introduces a rule-based adaptive background music system for the fighting game DareFightingICE, modulating volumes of five instruments playing Air on G-String based on in-game state cues (HP/EP of both players and inter-player distance). It evaluates the approach using a blind, audio-only DL AI across DareFightingICE versions 5.2 and 6.0, demonstrating performance gains over non-adaptive BGM. The study provides detailed mappings from game elements to instrument volumes and analyzes both objective results and qualitative AI behaviors, showing consistent improvements across sound designs. The authors plan to extend this with deep learning-based adaptation and include aesthetic evaluations to optimize BGM accordingly.

Abstract

This paper presents our work to enhance the background music (BGM) in DareFightingICE by adding an adaptive BGM. The adaptive BGM consists of five different instruments playing a classical music piece called "Air on G-String." The BGM adapts by changing the volume of the instruments. Each instrument is connected to a different element of the game. We then run experiments to evaluate the adaptive BGM by using a deep reinforcement learning AI that only uses audio as input (Blind DL AI). The results show that the performance of the Blind DL AI improves while playing with the adaptive BGM as compared to playing without the adaptive BGM.

Figures (4)

• Figure 1: Rule-Based Adaptive BGM.
• Figure 2: Adaptive BGM Rules
• Figure 3: Adaptive BGM Code Example
• Figure 4: A series of fight scenes at different time steps and the level of different instrument volume at these time steps.