Sound recognition ‘key’ in front-door tech

Chris Mitchell

Written by
Dr Chris Mitchell, CEO and Founder

March 7, 2018

Front-door tech is hot, according to TechCrunch. For those companies trying to make our homes more secure or e-commerce more streamlined, the ability to embed AI (via computer vision and sound recognition) provides opportunities to better equip consumers’ homes with the smarts to safeguard our parcels, property and possessions.

The market has been growing over the last few years and following the Ring acquisition last week, the August Home acquisition by Yale owner ASSA ABLOY in October, and the Nest Hello launch back in October there appears to a battle among the technology companies for our front porches.

Amazon (we assume) is interested as they expand their Amazon Key and Prime services and overcome that last great hurdle of online shopping – the delivery. According to the FT, 30% of US households have had a package stolen from outside their home. So being able to get packages into safe places has benefits for consumers, Amazon and the couriers.

But there are wider safety and security applications, especially at night or when the house is unoccupied. What is happening outside a property can be an important precursor to something undesirable, such as vandalism, theft or burglary.

For clarity, front-door tech to us is a general term covering smart locks, doorbells and outside security cameras.

Both cameras and smart doorbells utilise HD video, night vision, motion detection and in Nest’s case, person recognition but as with other aspects of the smart home, being able to hear as well as see gives our digital companions richer contextual information on which to assess threats. In turn, this helps us to respond when we need to.

So what are the sound recognition applications within front door tech?

  • Audio anomalies – sounds that occur during the day. Not necessarily triggering an alert but activating a video clip. Or alternatively, a hyper-vigilant sensor that is activated upon detecting other visual or audio activity and which captures the sounds and video as situations develop. Particularly useful if services such as Vivint’s Streety neighbourhood watch take-off.
  • Window glass break – whether detecting the sound of somebody breaking into a property or car, this detection could trigger a video clip and alert, or instruct other devices to respond (such as turning on lights, playing an alarm, etc).
  • Speech detection – not to be confused with speech recognition, this is detecting the presence of human voices which would indicate that people are approaching or loitering outside.
  • Car horn – was something happening in the street outside a consumer’s property that could have been related to a disturbance or issue?
  • Door knock/doorbell – did a visitor miss a consumer’s smart doorbell or does the consumer not have one? A security camera that can detect the sound of the traditional doorbell or door knock sound could alert the owner that somebody is at the door.
  • Door open/close – has the door been opened or closed? One indicates occupancy whereas the detection of the door closing could alert other smart home systems that the property should now be vigilant as the home is unoccupied. Or perhaps the fact that the sound of the door closing didn’t occur might alert the homeowner to the possibility that they didn’t close the door behind them.
  • Dog bark – specifically for couriers to be able to understand the likelihood of a dog being present within the house. As services like Amazon Key take off and homeowners can let a delivery driver into their home remotely or automatically it would be good for the courier to understand the risk of a dog escaping or attacking.

Hive’s Hub 360 makes the smart home smarter with our sound recognition technology

Read more

Sound recognition ‘key’ in front-door tech

Read more

What does the Ring acquisition mean for the smart home market?

Read more

Our software

Our embedded software platform, ai3™, is integrated into consumer products to make them more intelligent by understanding the sounds around them.

Find out more

Use cases

Our sound recognition software has been designed to work in a wide range of products from light bulbs, to smart speakers and beyond.

Read more