Improving multichannel speech enhancement through accurate room-acoustic simulations
For researchers in speech enhancement, this work demonstrates that simulation fidelity directly impacts real-world performance, though the approach is incremental.
The paper shows that using high-fidelity room-acoustic simulations for data augmentation reduces median word error rate by up to 38% compared to lower-fidelity simulations in multichannel speech enhancement.
Room-acoustic simulations are widely used to augment training data for deep-learning-based speech enhancement. While most pipelines rely on simplified geometrical acoustics, wave-based approaches offer greater physical accuracy. In this work, we examine how simulation fidelity affects multichannel speech enhancement performance. To this end, we train SpatialNet on datasets augmented with different room-acoustic simulation methods and evaluate the resulting models on measured data. We compare lower-fidelity datasets based on geometrical acoustics with a high-fidelity dataset using advanced acoustic modelling and a hybrid combination of wave-based and geometrical acoustics simulations. Training on the high-fidelity dataset results in an up to 38 % relative reduction in median word error rate compared to the lower-fidelity alternatives. These results show that augmentation with high-fidelity room-acoustic simulations directly translates into improved multichannel speech enhancement performance.