STEP Labs’ technology is about reducing the impact of noise on voice communications. Our first strength is in separating voice signals from non-voice signals like music, wind and noise, to let people and machines understand what is being spoken. And, because consumers in the real world need to hear as well as be heard, we developed intelligent speaker solutions that adjust output volume, and compensate in noisy environments by punching up the clarity of the sounds played through the speakers. All of STEP Labs’ technologies are software-based, always-on solutions, which require no additional battery power, no special physical touch-points or sensors, and can be placed anywhere near the person speaking – on an ear, in electronics, in a car, on a desk, in clothing… the options are endless. Hearing Is Believing In the example below, we used a commercial dual microphone headset from a major brand. We only replaced the DSP algorithm with STEPvoice, our powerful voice separation algorithm. The background noise consists of music and babble and is set at 83dB. You can hear by comparing the sound clips in our applications webpages, that STEPvoice greatly improves the SNR, while maintaining the voice quality. You can also see the difference in background noise with and without STEPvoice in the clips on our home page. A better SNR combined with an excellent voice quality allows clean and clear communication as well as near-perfect voice recognition performance. Expect More. Crystal clear and clean voice signal, free of noise and distortion for clear and reliable hands-free communications and near-perfect voice recognition performance, regardless of the applications and the environments. R&D by STEP Labs Start the revolution We invent. We don't use conventional noise reduction methods and their inner limitations. We invented a method that greatly respects the voice content and separates it from ambient noises. STEP's technology is based on a new and innovative approach. As a result, we achieve excellent SNR improvement even in extreme noise while still preserving the voice quality without any distortion or artifact. Therefore, we enable unmatched hands-free voice recognition performance.