Microchip on a finger

Solving the "moving target" challenge: proactivaudio and Infineon bring clear audio to the Philips SpeechMike Ambient Wearable AI Assistant

Internet of Things

In the rapidly evolving landscape of voice technology, the most advanced algorithms require equally capable hardware to reach their full potential. This reality is the foundation of the latest success story between Infineon and our Associated Partner, proactivaudio. By combining proactivaudio’s sophisticated audio signal processing with Infineon’s precise MEMS microphone technology, the collaboration has enabled Speech Processing Solutions, authorized Philips licensing partner for professional dictation solutions to launch their newest innovation: the Philips SpeechMike Ambient Wearable AI Assistant.

For professionals on the move, every minute counts. In a hectic day-to-day working environment, they are always on the go, facing multiple dynamic scenarios—from quiet office work to consulting in noisy wards. While documentation is essential, finding the time to sit and type is becoming impossible. They needed a solution that adapts to their movement, allowing them to document instantly without being tethered to a desk.

The new Philips SpeechMike Ambient is designed to revolutionize voice-driven documentation across industries, through proactivaudio's engineering, the device now supports wearable, docked, or on-table hands-free recording, giving users complete freedom of movement.

PSM5000 Philips
PSM5000 Philips
PSM5000 Philips

To achieve this flexibility, proactivaudio deployed a sophisticated software stack featuring:

  • Adaptive beamforming: This technology addresses multiple use cases and scenarios, ensuring the microphone array adapts instantly whether the speaker, e.g. the doctor, is sitting, standing, or moving.
  • Advanced Voice Processing: The system enables conversational recording and the removal of ambient voices in single user mode, ensuring greater accuracy in dynamic, noisy clinical environments — far outperforming smartphones or generic microphones.

Efficiency: The solution utilizes lightweight firmware, which allows for up to 10 hours of recording in a single battery charge—a critical feature for long shifts.

To execute these complex algorithms effectively, particularly the adaptive beamforming, the hardware foundation had to be flawless. proactivaudio recommended and designed-in Infineon’s IM69D120 XENSIV™ MEMS microphones to meet the strict requirements of their array processing.

For the algorithms to function with high precision, the microphones needed to exhibit exceptionally low tolerance variations and a high Signal-to-Noise Ratio (SNR). Infineon’s MEMS technology provided the necessary consistency and acoustic performance to turn proactivaudio’s complex acoustic software concepts into a working reality embedded in a compact, battery-operated device.

The synergy between proactivaudio’s intelligent software and Infineon’s robust hardware has resulted in a device that ensures "perfect recording quality" for speech recognition. Philips’ latest product highlights the added value of the Infineon Partner Ecosystem overcoming design challenges: By connecting specialized software experts like proactivaudio with the right hardware foundation, we empower customers like Philips to master complex engineering challenges and accelerate the development of next-generation audio applications. Innovation that helps doctors where they need it the most. 

proactivaudio logo
proactivaudio logo
proactivaudio logo

proactivaudio is a company specializing in ML-driven, real-time audio processing for speech interfaces and communications. Their solutions combine noise reduction, echo cancellation, and speech enhancement to outperform traditional DSP, solving echo/noise, source separation, and IoT speech interface needs.

Preferred Partner
Preferred Partner
Preferred Partner