This is how HomePod hears you even with background noise

The HomePod now It has landed in various countries. Although there has not been much information about it, it is known that 600.000 units were sold in the first quarter. Apparently this number is slightly lower than Apple expected. However, the great work of engineering in which large amounts of money have been invested, is a work of art.

A work of art not only on the outside. So is all the mechanics that it carries inside: from the A8 chip that governs the entire structure to the different types of high-quality speakers through the artificial intelligence used. Today, Apple engineers have solved an unknown: how Siri hears you even with background noise.

Siri, HomePod and artificial intelligence: a perfect trio

The Apple-designed A8 chip is the brain behind the HomePod's great innovations. Like advanced signal processing so Siri can hear you even when the music is playing. Or studio-quality real-time processing, which boosts bass and reduces distortion. Also, buffering is even faster than real-time. And the combination of live and ambient audio gives you unrepeatable sound, time after time.

At the HomePod presentation more than a year ago, we were assured that Siri would respond to our questions even with background noise. Since its commercialization, tens of thousands of people have been able to verify it satisfactorily. It doesn't matter if the music is blasting or there is a conversation in the room, the moment you say "Hey Siri," HomePod mutes and gives you the information you're looking for.

The engineers who are part of the Audio Engineering team and the Siri team have explained in an article how it works in a specific way by which Siri can hear us even with background noise. As explained, the HomePod can hear you in three situations: when there is music, when you are talking to it from a distant place or when another sound such as television is interfering. The key to everything is in a multichannel processing system based on two premises:

  • Echo and background noise removal mask using artificial intelligence
  • Determining which audio fragments to remove from which to keep

The speaker employ six microphones and they work at the same time continuously to provide information to the multichannel system. This is important as such a task would consume a lot of energy. However, this process only occupies slightly more than 10% of the capacity of the A8 chip. If you want to know some examples of how the multichannel system improves, you can access the blog from the engineering team where there are several examples to prove it.


Follow us on Google News

Leave a Comment

Your email address will not be published. Required fields are marked with *

*

*

  1. Responsible for the data: AB Internet Networks 2008 SL
  2. Purpose of the data: Control SPAM, comment management.
  3. Legitimation: Your consent
  4. Communication of the data: The data will not be communicated to third parties except by legal obligation.
  5. Data storage: Database hosted by Occentus Networks (EU)
  6. Rights: At any time you can limit, recover and delete your information.