microASR: 32-μW Real-Time Automatic Speech Recognition Chip featuring a Bio-Inspired Neuron Model and Digital SRAM-based Compute-In-Memory Hardware.
Dewei WangJonghyun OhGregory K. ChenPhil C. KnagRam K. KrishnamurthyMingoo SeokPublished in: ESSCIRC (2023)
Keyphrases
- bio inspired
- automatic speech recognition
- real time
- evolvable hardware
- low cost
- neuron model
- random access memory
- speech recognition
- swarm intelligence
- low power
- speech signal
- hidden markov models
- artificial neural networks
- memory access
- memory subsystem
- power consumption
- feed forward
- spiking neurons
- address event representation
- image processing
- data mining
- vision system
- control system
- feed forward neural networks
- biologically plausible
- power dissipation
- language model
- evolutionary algorithm
- artificial intelligence
- genetic algorithm
- ibm power processor