- The active noise cancellation feature on Skype is designed to silence everything but the user's speech.
- Currently available to Mac and Windows, Skype users can tweak the noise cancellation settings between low, medium, and high.
- Skype notes the noise cancellation is supported by neural networks that can learn the difference between speech and more complex, inconsistent noises.
Skype for Desktop has introduced the noise cancellation feature for more efficient video calls. The company said in a blog post that the noise cancellation feature works by using specially trained deep neural networks that filter out the noise without affecting the speaker's voice. The feature was originally developed for Microsoft Teams and has now been extended to Skype. "We are pleased to announce the release of our latest background noise suppression feature in the Skype desktop app. Originally developed for Microsoft Teams, this new feature is designed to silence just about everything except for your voice when you're meeting on Skype," Skype noted. The noise cancellation is currently only available on Windows and Mac on Skype for the Desktop app.
Users can access active noise cancellation on Skype by following the given steps:
-- Login to your Skype account.
-- Go to Profile > Settings > Audio and Video.
-- Go to the Audio tab.
-- Select the Noise cancellation option.
-- Select from Auto, Low or High noise cancellation.
-- You can also turn the toggle off if you want to disable it.
Skype in the blog post explains that the neural networks which support the noise cancellation can learn the difference between speech and more complex, inconsistent noises such as typing on a keyboard, the crunch of a food wrapper, and a howling dog, while traditional noise suppression algorithms can address only simple, consistent noises like that of a fan.
"This technology relies on machine learning (ML) to learn the difference between clean speech and noise and is frequently referred to as artificial intelligence (AI). A representative dataset is used to train the ML model to work in most of the situations our Skype users experience. There needs to be enough diversity in the dataset in terms of the clean speech, noise types, and the environments from which our users are joining online calls," Skype noted.
It further noted that it did not use customer data for the datasheet but the one that was available in public.
Skype noted that for clean speech sounds, the company used a balance of female and male speech and data from over languages, including tonal languages. It further noted that for noise data, Skype included 150 noise types to cover the diverse scenarios our users may run into from keyboard typing to running water, to snoring. It also included emotions in the data so they do not get suppressed while noise cancellation.