Automatic speaker recognition (ASR) is a very difficult problem when applied to noisy environments. Even sophisticated ASR sytems such as Alexa or Siri will fail if there is too much noise in the background. Those handy little helpers will not be able to transcribe speech properly, leading to a negative user experience when used in cars, offices or smart-homes.
Normally, you don't want any negative feelings linked to your product, but outstanding user experience and handy state-of-the-art technology instead. Evolve is able to solve this problem for audio and ASR. Our algorithm verifiable removes unwanted background noise and preserves the speech quality. Not believing it? For demonstration purpose, we prepared two audio recordings while driving a car. The first one contains the original recorded sound. You will hear someone talking and some music in the background. In addition, there is also some noise coming from the engine of the car.
The second recording contains the same data, but is processed with our powerful noise cancelation software. The algorithm is running locally on a tiny chip inside the car, without the need of any internet. Do you hear the difference? Our algorithm significantly improves the perceptual audio quality and verifiably boosts the speech recognition (ASR) acurracy when using driver assistant systems. This enables high quality ASR in car environments and leads to a unique listening experience, never achieved before.
Please drop us an email if you want to integrate our system into your business process.