Why do I keep doing this

Use a combination of different size microphones diaphragms and not just MEMS devices, Use the earshape parabolic concentration points for finding the optimal location and orientation of diaphragms include ultrasonic transceivers for out of band perception (also echolocation, room mapping, darksight) also to allow covert/out of band ultrasound communication with other catgirls that normies cannot hear (that means connecting transceivers AS transceivers, with US band amplifier for transmission) Using servogalvos to modify ear shape for acoustic focusing, if ear direction is using a gear drive, use double helical gear and a low acoustic profile dampening coating on the gears (acoustic anti-reflection) because we are making a listening device and we don't want to listen to gear noises ! Consider ultrasonic motors instead of geared drive. Software wise, using direct wifi and ble for communication (DECT sucks, proprietary standard sucks) this should work with a raspberry pi and only open source software out of the box, no drivers needed, yes that means multichannel transmission and do all the processing off device actual software, we're using the various microphones as an asynchronous, heterogenous phased array, unlike regular phased arrays which have many identifical receivers along a straight plane here we are going to skip a couple technological steps and use "receiver diversity" in the phased array design, so not only are the receivers in different places NOT on a plane, they are also receivers with different acoustic profiles (different sensibility of different frequencies) and anisotropic (due to ear "lens-shape") reception patterns, so all of that diversity has to be correlated between all receivers to create an acoustic map of the world, so that we can decide also what sounds are psychoacoustically relevant and point the ears in that direction, listen if the received signal is actually signal or noise and then send that to the user, in realtime (max processing time is 20ms top end) make sure ear shape flexibility is enough to receive and focus some signal backwards.. there's still a few things to consider but I think that's mostly it, for the earbud, use close loop ear canal acoustic feedback for the amplifier drive, use only open standards and open firmware in all devices. System to use myoelectic ear movement as control input feedback, at least with these two modes, ear movement to manual steering of the acoustic phased array, and the other ear movement to select auditory focus of signal sources (that is, automatic switching between detected relevant auditory sources). Add RED+IR laser pointer to the ears that can be used as a pointed, to indicate (covertly for IR) which auditory source is currently select and in what direction this signal currently use (ensure positional live tracking is reflected in beam direction). Use sub-vocal command interface for higher function control of the listening device. Provide standardized pattern and calibration equipment definition for microphone correlation calibration, use only open hardware assemblages. Include live transcription to text of all received text, ensure at least capacity for 12 simultaneous audio to text transcription channel, include live audio MP3QR decoding and all other digital audio transmission standard live decoding and transcription as part of the transcription channel capability. Include text summary ability, include voice signature automatic identification and tagging of decoded text stream, at least 4 terabyte storage space, maximum head weight 150 gram excluding battery, battery system must have 1 minute with no battery operation time or dual battery swapping system. include hardware audio encoder and decoder supporting all current ffmpeg codecs. Include acoustic counter battery targetting receiver system. Oh almost forgot, dual laser system mounted with microprism(for aimed laser microphone ability, to listen sounds through walls or from vibrating surfaces)

Ok Machine, convert this schizo-ing document into Ratheon Cybernetic Auditory Telemetry & Echolocation with Anisotropic Reception (CAT&EAR) engineering requirement draft document