🤖 AI Summary
This work addresses the performance limitations of traditional adaptive room equalization methods, such as the filtered-x least mean squares (Fx-LMS) algorithm, which suffer from structural rigidity in time-varying acoustic environments and under complex excitations like music. The authors propose a modular, differentiable digital signal processing (DDSP) framework that, for the first time, integrates classical adaptive filtering with DDSP. By leveraging automatic differentiation, the Fx-LMS algorithm is embedded within a unified formulation, enabling flexible substitution of equalization structures, room response estimators, loss functions, and optimizers. The framework reveals that frequency-domain objectives yield superior adaptation stability in time-varying scenarios. Experimental results demonstrate up to a 70% reduction in system distance and a 13% decrease in Mel-spectral distance, while elucidating the trade-off between room response estimation accuracy, frame length, convergence stability, and adaptation speed.
📝 Abstract
Adaptive room equalization remains challenging under time-varying acoustic conditions and complex excitation signals, such as music. In these scenarios, classical filtered-x least mean squares (Fx-LMS) methods falter due to their rigid formulation. We present a modular differentiable digital signal processing (DDSP) framework for closed-loop adaptive room equalization that recovers Fx-LMS as a special case through automatic differentiation. The framework supports interchangeable EQ structures, response estimation methods, loss functions, and optimizers. Experiments with time-varying measured room impulse responses show that frequency-domain objectives provide more stable adaptation than time-domain objectives in the considered scenarios. Relative to the non-equalized response, system distance is reduced by 70% and mel-spectral distance by 13% (worst-case scenario). We further examine how online room response estimation accuracy and frame length affect the trade-off between responsiveness and convergence stability. Overall, the framework provides a unified open-source basis for exploring synergies between classical adaptive filtering and DDSP-based optimization.