Hybrid Neural Networks for On-device Directional Hearing
This addresses the problem of low-latency audio source separation for computationally-constrained wearables, though it is incremental as it builds on existing methods.
The paper tackled on-device directional hearing by developing a hybrid model that combines traditional beamformers with a lightweight neural net, achieving comparable performance to state-of-the-art models with a 5x reduction in model size, 4x reduction in computation per second, and 5x reduction in processing time.
On-device directional hearing requires audio source separation from a given direction while achieving stringent human-imperceptible latency requirements. While neural nets can achieve significantly better performance than traditional beamformers, all existing models fall short of supporting low-latency causal inference on computationally-constrained wearables. We present DeepBeam, a hybrid model that combines traditional beamformers with a custom lightweight neural net. The former reduces the computational burden of the latter and also improves its generalizability, while the latter is designed to further reduce the memory and computational overhead to enable real-time and low-latency operations. Our evaluation shows comparable performance to state-of-the-art causal inference models on synthetic data while achieving a 5x reduction of model size, 4x reduction of computation per second, 5x reduction in processing time and generalizing better to real hardware data. Further, our real-time hybrid model runs in 8 ms on mobile CPUs designed for low-power wearable devices and achieves an end-to-end latency of 17.5 ms.