Overview

We introduce a new online adaptive filtering method called supervised multi-step adaptive filters (SMS-AF). Our method uses neural networks to control or optimize linear multi-delay or multi-channel frequency-domain filters and can flexibly scale-up performance at the cost of increased compute – a property rarely addressed in the AF literature, but critical for many applications. To do so, we extend recent work with a series of improvements including feature pruning, a supervised loss, and multiple optimization steps per time-frame. These improvements work together in a cohesive manner to unlock scaling. Furthermore, we show how our method relates to Kalman filtering and meta-adaptive filtering, making it seamlessly applicable to a diverse set of AF tasks. We evaluate our method on acoustic echo cancellation (AEC) and multi-channel speech enhancement tasks and compare against several baselines on standard synthetic and real-world datasets. Results show our method performance scales with inference cost and model capacity, yields multi-dB performance gains for both tasks, and is real-time capable on a single CPU core.

We have released code and will release model checkpoints using the metaaf python package developed for this work. For demos of the code, setup instructions, and more, check out the GitHub repo. To listen to model outputs, keep scrolling.

We release model inputs and outputs for both the acoustic echo cancellation (AEC) and generalized sidelobe canceller (GSC) experiments.

AEC Demo Files