Piezoelectric MEMS Microphones: Build Reliable Arrays for Far-field Voice Recognition
Voice Assistants are a fast-growing product category in the Consumer Electronics industry. Adoption of voice assistants is increasing year-over-year with integration into smart speakers, mobile phones, wearable devices, etc. Improvement in speech recognition accuracy obtained from advancements in Natural language processing, machine learning, and cloud computing technologies is driving the success of voice assistants. But, the true potential of these assistants is still limited by the quality of the input signal captured by the microphone at the front end of the audio signal processing chain. This paper provides insight into microphone selection criteria for far-field performance using the measurement data taken with piezoelectric microphones in comparison to standard Electret Condenser microphones (ECM). Test results on far-field performance metrics obtained at an independent test laboratory are analyzed with respect to the microphone characteristics.