Imagine a scenario where you are not at home and a burglar breaks into your house and your dog starts to bark. What if AI could read the dog’s bark and transfer it to your device where immediate action can be taken? Sound is an excellent medium when it comes to communication. But while technology has made great advances in the realm of visual and touch cognition, the full potential of sound has not yet been explored. In fact, many devices offer ‘speech’ recognition but not all types of ‘sound’.
Korean startup Cochlear.ai has ventured into the domain by delivering top-quality machine listening technology to solve issues and challenges around the world. The technology can empower developers to make creative applications using audio cognition technology without a deep understanding of audio processing. The company uses cloud API and Edge SDKs which adds hearing abilities to any device or applications.
Simply put, the technology recognizes ‘sounds’ such as bark, a scream, a laugh, baby’s cry, sound of a glass breaking and transfers it to a device or application where.
Next-gen intelligence, real-time analysis
Clues of emergency are often clearer in sounds, such as screaming or glass breaking, and these events can be detected by the machine listening AI. The automated monitoring system is more efficient and cheaper than human labor when it comes to eradicating human errors. The technology can be used for content-based music recommendation or searching and grouping huge quantities of music clips.
The technology aims to recognize the message beyond the speech text and grasp the meaning behind the tone or intensity of the sound in order to detect illness, happiness, cry for help, etc. It can also be used to accurately detect anomalies in the machine by the sound being produced when the machine malfunctions.
What’s more, the technology will help search form information using audio clips, something which is limited to text and image searches currently.
Appreciation of the cutting-edge sound AI research
The Cochlear.ai team has already made waves! It achieved top ranks in all tasks of the IEEE Detection and Classification of Acoustic Scenes and Events (DCASE) Challenge 2017, the most prestigious competition in the machine listening field. In 2018, on the DCASE General Purpose audio tagging, the team ranked first among 558 teams on Kaggle.
Cochelar.ai’s team will be attending the GITEX TECHNOLOGY WEEK to be held from October 6 to 10 in Dubai, supported by the Korea Institute of Startups & Entrepreneurship Development (KISED). The GITEX is the biggest tech show in Middle East, North Africa and South Asia that attracts many prominent leaders from global companies, major media outlets, and influential investors.
… [Trackback]
[…] Find More on on that Topic: koreatechdesk.com/korean-startup-cochlear-ai-developing-audio-cognition-technology-to-understand-sound/ […]