Efficient and reconfigurable reservoir computing to realize alphabet pronunciation recognition based on processing-in-memory

With its high energy efficiency and ultra-high speed, processing-in-memory (PIM) technology is promising to enable performance in Reservoir Computing (RC) systems. In this work, we demonstrate an RC system based on as-fabricated PIM chip platform. The extracts input into a high-dimensional space through the nonlinear characteristic randomly connected reservoir states inside PIM-based RC. To examine system, dynamic predictions, including auto-regressive moving average equation of order 10 driven time series, isolated spoken digit recognition task, alphabet pronunciation, are carried out. saves about 50% requires much fewer operations as compared with implemented digital logic. This paves pathway for algorithm application lower power consumption less hardware resource required.