YouTube working over new AI noise cancellation feature

What is Noise cancellation?

The noise-canceling circuit senses outside noise with built-in microphones and sends an equal-but-opposite canceling signal to the headset. By producing this countersignal, your headphones block a large part of external sound sources. This is called sound compensation.

How does sound compensation work?

Did you know that Noise-Cancellation works independently for the Left and the Right channel? Sound compensation works predictive, providing inverse signals to the speakers who will then eliminate the noise. Because it is a predictive system, rapidly changing sounds or air pressures are compensated differently depending on how they reach the right or left noise-cancellation microphone. In other words, sound compensation is customized and can impact the output of each independent side.

What can I expect from Noise cancellation technology?

Noise-cancellation offers an enhanced listening experience by canceling out interfering with external sounds.

However, remember the following: 

  • The noise-canceling effect may not be pronounced in a tranquil environment, or some sound may be heard. If there is little ambient noise, it is even advisable not to switch on the NC function, as it needs ambient input to function optimally.
  • Noise cancellation doesn't transfer during telephone calls. Although it protects the headphones user from the surrounding noise, it does not apply to the receiver of the call: they will potentially hear background noise.  
  • Depending on how you wear the headset, the noise-canceling effect may vary, or a beeping sound (howling) may occur. In these cases, take off the headset and put it on again.
  • The noise-canceling function works for noise in the low-frequency band primarily (train, airplane, engine noise). Although noise is reduced, it is not canceled altogether.
  • When you use the headset on a train or a car, noise may occur depending on environmental conditions. It is not advisable to wear noise-cancellation headphones while driving, as taking notice of traffic sounds is essential for your safety.
  • Mobile phones may cause interference and noise. Should this occur, locate the headset further away from the mobile phone.
  • Do not cover the microphones. The noise-canceling function or the Ambient Sound Mode may not work optimally, or a beeping sound (howling) may occur. In these cases, take your hands off the headset microphones.

It allows the technology to start speech enhancement while the video is still being recorded. Those improvements shrunk the feature’s size from 120MB to 6MB, making it easier to deploy.

 You may notice a marked improvement in the audio quality of some YouTube Stories going forward, thanks to a new speech enhancement feature Google rolled out. Now, it’s making the technology available to creators recording YouTube Stories on iOS devices.

Modern active noise control is generally achieved through the use of analog circuits or digital signal processing. Adaptive algorithms are designed to analyze the waveform of the background aural or monaural noise, then based on the specific algorithm generate a signal that will either phase shift or invert the polarity of the original signal. This inverted signal (in antiphase) is then amplified and a transducer creates a sound wave directly proportional to the amplitude of the original waveform, creating destructive interference. This effectively reduces the volume of the perceivable noise. Active noise reduction (ANR), is a method for reducing unwanted sound by the addition of a second sound specifically designed to cancel the first.

Google says they also “reduced [Looking to Listen’s] running time from 10x real-time on a desktop using the original formulation... to 0.5x real-time performance using only an iPhone CPU.” How it has improved the technology over the past couple of years. To start with, the developers made sure that it can do all the processing on the device itself, so it doesn’t need to send anything to a remote server.

To ensure that it will work for everyone and won’t show bias, Google conducted a series of tests exploring its performance based on various visual and auditory attributes. Those attributes include the speaker’s age, skin tone, spoken language, voice pitch, visibility of their face, head pose, facial hair, presence of glasses, and the level of background noise. Facial hair doesn’t seem to have a big effect on it either, though it works best on faces with no facial hair and those with a close shave.


  1. Cancells 40+ noise types
  2. Stationary noises like a fan, AC, etc.
  3. Non-stationary noises like highway, train, wind, babble noises
  4. Highly dynamic noises like traffic horns, baby cry, police siren, dog bark, keyboard Clicks.


  • Low voice spectral distortion for HMI applications
  • Preserved Speech Intelligibility for voice calls
  • Platform-based customizations

I'm a Computer Science graduate, likes to do ordinary work in an extraordinary manner. I'm quite creative, a workaholic. I regularly used analyze new research, development, innovation by tech giants. I'm interested in Machine learning, Data Science along with research work applications on them & solving puzzles, quizzes.


Leave a reply

Please Login or Register to Comment. Get Started

Share this article