Spoken digit classification using a spin-wave delay-line active-ring reservoir computing
This work demonstrates a novel hardware implementation for reservoir computing, potentially benefiting neuromorphic computing and signal processing applications, though it appears incremental as it applies an existing method to a new physical system.
The researchers tackled spoken digit recognition using a spin-wave delay-line active-ring reservoir computer, achieving up to 93% classification accuracy, and also tested it on short-term memory and parity check tasks with capacities of 4.77 and 1.47, respectively.
As a test of general applicability, we use the recently proposed spin-wave delay line active-ring reservoir computer to perform the spoken digit recognition task. On this, classification accuracies of up to 93% are achieved. The tested device prototype employs improved spin wave transducers (antennas). Therefore, in addition, we also let the computer complete the short-term memory (STM) task and the parity check (PC) tasks, because the fading memory and nonlinearity are essential to reservoir computing performance. The resulting STM and PC capacities reach maximum values of 4.77 and 1.47 respectively.