
Unleashing a wave of smartphone innovation
Empowered by a compact sense of hearing.
Our Acoustic Scene Recognition technology enables smartphones to recognise the acoustic world around the user and react intelligently to chaotic, lively, boring, or calm soundscapes.
As this runs in a low-power, always-on mode on the new Qualcomm® Snapdragon® 778G Mobile Platform there are many things that smartphone OEMs can do.
For example, dynamically change notifications and alerts based on what is happening around the user so that they don’t miss an important call or message, or get caught in an embarrassing situation.
Our smartphones capture so much of the world around us, enabling us to relive or share precious memories with our friends, family, and should you want to, the rest of the world.
By automatically tagging the audio in consumers’ videos and photos, their content has never been more searchable. Plus, Media Tagging also makes it easier to edit and share video and photo content on the fly, helping users to express their creativity and personality.
Inject spontaneous AR fun into videos and video chat by using filters, frames and effects triggered by the sounds we make. Bring that long-distance bedtime story to life with a well-placed, authentic lion roar. Keep family, friends or total strangers entertained on social media.
By embedding our sound recognition software into your camera, messaging or video chat application you can allow consumers to trigger a wide range of effects based on their ability to quack, meow, moo and baa and more.
And by using sound recognition, users can let the AI respond to the sounds they make, allowing the conversation to flow more naturally.
Support a wide range of users with disabilities and impairments through a sound-based accessibility feature that alerts users who are deaf, have partial hearing loss or are wearing headphones to important sounds within their environment.
Whether that is a smoke alarm, baby crying, dog barking or an impatient delivery driver looking to knock and run, a smartphone running our reliable sound recognition software can augment the world around the user to keep them safe and sound.
56x better classification performance than iOS14
In our field tests, Apple’s iOS14 Sound Recognition Accessibility feature produced at least 56 times more false positives than we have seen in commercial deployments of our technology.
Minimal impact on battery expectations
Our ai3-nano™ software consumes the same amount of power as a wake word module. As a comparison, Apple’s iOS14 Sound Recognition feature is power-hungry, consuming 46% of the iPhone’s battery over the course of a day in our field tests.
So compact it runs concurrently with voice assistant
The compactness of our software means that, for the first time, users don’t need to decide between sound recognition and speech recognition when using our technology, which means they benefit from both core AI features.