Cochl.Sense

What is Cochl.Sense?

Cochl.Sense is a machine-listening platform. It turns audio into structured insight — the sounds present, the speech being spoken, the scene as a whole — and runs anywhere your audio does, on cloud servers or a wide range of edge devices, including microcontrollers.

What can Cochl.Sense do?

Cochl.Sense exposes three audio-understanding capabilities:

• Sound Event Detection — recognize what sounds are present (sirens, glass break, baby cry, and 100+ tags). Available via the Cloud API, the Edge SDK, and Cochl.Sense Nano on microcontrollers.
• Speech Analysis — transcribe speech and identify registered speakers. Available on the Cloud API only.
• Audio Insights — high-level summary of an audio scene (environment, situation, keywords). Available on the Cloud API only.
• Custom Sound — extend the catalog: Sound Events on the Edge SDK, or Speaker Profiles for Speech Analysis on the Cloud API.

How do I use Cochl.Sense?

Cochl.Sense comes as three products, split by where the model runs. Two of them you can start on your own: Cochl.Sense Cloud API and Cochl.Sense Edge SDK.

Cochl.Sense Cloud API

Send audio (WAV, MP3, FLAC, OGG) to detect sounds over HTTP.

Cochl.Sense Edge SDK

Run it on edge devices using Python, C++, or Android.