What is Cochl.Sense?

Our daily lives are surrounded with millions of different sounds that contain important clues of our environments. For instance, if you hear someone screaming, you know there is an emergency situation. If you hear a siren, you know an emergency vehicle is approaching. Humans are good at understanding sounds and connecting them to proper contexts. Machines, on the other hand, have no abilities to understand sounds even though they are much more powerful when it comes to hearing sounds. How great would it be if AI could let machines understand sounds?

Cochl.Sense allows computers to understand sounds with machine listening technologies. With advanced audio processing and neural network techniques, it provides noise-robust and environment-independent machine listening. Just input your audio data (either file or stream) into Cochl.Sense, and it will detect and return what kind of the input sound is. It can be used on any device, any platform (smart speakers, IP cameras, etc.).


What does Cochl.Sense detect?

The Cochl.Sense can detect different events (e.g., Siren, Gunshot, Scream). You can see a more precise list during the project creation on our Dashboard.

How do I use Cochl.Sense?

You can use Cochl.Sense in two different ways: Cochl.Sense API and Cochl.Sense SDK.


Cochl.Sense API

Send prerecorded audio (codecs supported: WAV, MP3, FLAC) or real-time streaming to Cochl.Sense API platform, and it will detect and return what that sound is.
Cochl.Sense API can be used on any environment HTTP is supported.


Cochl.Sense SDK

With Cochl.Sense SDK, Cochl.Sense can be integrated to your program and run on edge devices.
Cochl.Sense SDK provides C++, Python language on Linux platform.