Career:The Adaptive Silicon Cochlea: Biology, VLSI, and Applications

点击翻译按钮获取中文摘要

We described a 117-stage 100Hz-to-10kHz silicon cochlea that attained a dynamic rangeof 61dB while dissipating 0.5mW of power. This cochlea has the widest dynamic rangeof any artificial cochlea built to date. The wide dynamic range was attained in a low-noise traveling-wave architecture with distributed gain control. The analysis of ourelectronic cochlea suggested why nature preferred a traveling-wave mechanism over abank-of bandpass filters to decompose sounds.We propose to use the silicon cochlea in 3 projects that will range over a span of 4-5years. The first project, Fundamental Issues in Cochlear Gain Control will extend andexpand our previous work on gain control in the cochlea to improve its performancefurther, will help understand the functional role of gain control in the biological cochleabetter, and will yield computational primitives and circuits that are useful for cochlear-implant speech processors, and for robust-and-adaptive speech and auditory-patternrecognition. We expect this project to be funded soley through the CAREER proposal.The second project, Cochlear-Implant Speech Processors for the Deaf will focuss onusing the silicon cochlea as a low-power front end in cochlear-implant speech processors.We expect the performance of these processors to significantly improve because thesilicon cochlea models several important effects that are exhibited by the biologicalcochlea that current implants do not, and also because it can implement thesecomputationally-intensive biological algorithms in real time and with low power. Thecochlear-implant project will be done in collaboration with Dr. Donald Eddington at theMassachusetts Eye and Ear Infirmary, who is a pioneer in the field of cochlear implantsand who is very happy to collaborate with us. We expect to use CAREER funding to helpus get started on this project, but fund it primarily via other grants from the NIH or NSF.The third project, Adaptive Front Ends for Speech and Pattern Recognition will focuss onusing spike-based hybrid (analog-digital) computation techniques that were invented bythe author at Bell Labs to construct an energy-efficient and adaptive trellis vectorquantizer, an important component of speech preprocessors. The trellis vector quantizeris useful in quantizing sequences of analog vectors with certain temporal dependenciesbetween them, such as those that emerge from a cochlea; we plan to implement onchiplearning to adapt the parameters of this quantizer; the architecture of such a quantizermaps naturally to the architecture of a hybrid state machine (HSM), a machine thatextends and expands the concept of a finite state machine to the hybrid domain. Weexpect to supplement the beginning funding for this project from the CAREER proposalwith funding from Lucent Technologies or other funding agencies. We will collaboratewith speech researchers at Bell Labs and at MIT on this project.We propose to develop two courses at MIT, entitled Feedback in Electronics andBiology, and Hybrid Computation over a span of 3-4 years that will introduce intoelectronics, neural examples and concepts in adaptive-and-hybrid computing.