Tech Updates

Google Meet Introduces AI and Machine Learning-Based Noise Cancellation Feature

The immediate demand for Internet-based meetings has risen for a few months due to the increased number of employees working from home. The Covid-19 pandemic has shut down all the organizations and forced employees to work from home. To help everyone transition to a home office environment, Google is updating its premium virtual meeting app, Google Meet, with the much-required noise cancellation capability. The feature uses Artificial Intelligence (AI) and Machine Learning algorithms to eliminate noise in the background for video calls/conferences. What’s more? The feature leverages supervised learnings that necessitate training an AI model in a labeled data set and comes handy in improving the quality of group meetings. Besides, the AI-enabled noise cancellation hit the web first, following with the iOS and Android coming later.

Here’s How It Works

The feature announced later in April for G Suite Enterprise and G Suite Enterprise for Education customers, but since the rollout continued, it is now becoming accessible for Meet users on the web. However, the noise cancellation gets activated by default but can be turned off by heading over to the audio settings and turning the noise cancellation toggle off.

Serge Lachapelle – G Suite’s Director of Product Management says “The capability operates with the help of machine learning algorithms. The audio of the speaker is transmitted securely to the cloud where a machine learning algorithm analyzes it to identify the main voice and remove all the ambient noises surrounding it in real-time”. However, he adds “Once the processing takes place, the filtered out voice - free from ambient noise - is delivered to the person on the other end of the call. Overall, the quality of the voice comes out pretty clear, but the segments where the processing has been done sound a little degraded or low. But, it will improve eventually.

Serge Lachapelle – G Suite’s Director of Product Management says “The capability operates with the help of machine learning algorithms. The audio of the speaker is transmitted securely to the cloud where a machine learning algorithm analyzes it to identify the main voice and remove all the ambient noises surrounding it in real-time”. He further continued “Once the processing takes place, the filtered out voice independent from ambient noise is delivered to the person on the other end of the call. Overall, the quality of the voice comes out pretty clear, but the sections where the processing was been done sounded a little deprived or low. But, it will improve with time".

Curious already? Here is the video featuring Serge Lachapelle- G Suite’s Director of Product Management. In this video, he is describing the working of “Denoiser” in detail using a snack bag, a clicking pen, and rotating a metal object in a glass.

Get In Touch

Ask Us Anything !

Do you have experience in building apps and software?

What technologies do you use to develop apps and software?

How do you guys handle off-shore projects?

What about post delivery support?