Watch our Demonstration Video

Ambient sound is an underutilized resource for many industrial, automotive and consumer applications. Reality AI is highly effective at extracting signatures of specific events and conditions in noisy environments. 

Machine noise

Many machine conditions make specific sounds.  

If a technician can hear when something is wrong, Reality AI can be trained to hear it too.

  • Automotive – engine, tire, alignment and other mechanical issues

  • Valves and pumps – sounds associated with different flow conditions

  • Compressors, condensers and motors

Automotive uses

Automotive suppliers and OEMs are starting to make use of sound for a variety of purposes:

  • Sensing road conditions and surrounding environment in support of autonomous operation

  • Identifying events and conditions related to human occupants in the cabin

  • Predictive maintenance and condition monitoring beyond whats available on the OBD

Consumer products

As voice control becomes more prevalent, microphones are finding their way onto consumer devices. 

Ambient background noise and signal properties of sounds can give important context:

  • Ensuring wake words and spoken passwords come from live people and not recorded playback

  • Identifying background noise that gives clues to what's going on (food cooking, tv on, etc)

  • Listening for sounds of device or machine malfunction

Note:  Reality AI is not suitable for voice recognition or natural language processing.  Other tools will give you better results for these applications.

Learn more with our Technical Whitepaper

Explore the technical details behind the Reality AI approach to machine learning with signals

  • Why signals require a different approach than other machine learning problems

  • The importance of "features" to effective machine learning

  • Why the FFT probably isn't good enough, and what other options are better

  • The difference between Reality AI and Deep Learning