08:45 – 09:05

Keynote

How will Generative AI shape the future of MEMS Microphones?

In the present era, the utilization of AI in diverse applications is steadily increasing, contributing to strength individuals’ efficiency. Tools like ChatGPT play a pivotal role in enhancing people productivity by engaging users through text or voice inputs. Notably, Infineon’s high-performance silicon microphones enable seamless speech interaction with Natural Language Processing assistants. Prior to being processed by Large Language Models, audio data undergoes Speech-to-Text (STT) conversion, a crucial task to ensure accurate input for AI systems. STT operates by breaking down speech into “phonemes” and utilizing spectral comparison to transcribe the audio. Infineon’s initial trials evaluate the significance of high Signal-to-Noise Ratio (SNR) in applications like laptops interpreting the Error Word Rate. Furthermore, in True Wireless Stereo (TWS) and Smart-VR glasses devices, the fusion of AI with Infineon’s Vibration sensor shows the potential to revolutionize current architectures, delivering clear phone calls even any environmental conditions and outperforming current solutions.

Aldo Bruno

Technical Lead MEMS Sensor System Competence Team

Infineon Technologies AG