Getting Started

Cochl.Sense Edge SDK is a powerful toolkit that lets you seamlessly integrate machine listening into your applications and run it directly on edge devices. It enables low-latency, offline sound recognition without relying on the cloud, making it ideal for scenarios where real-time response and data privacy are critical.

1. Prepare Your Environment

Prepare your environment for the Cochl.Sense Edge SDK.

(1) List of supported platforms

  • Linux Desktop (Ubuntu 18.04-24.04)

  • Single-board Computers:

    • NVIDIA Jetson AGX Xavier
    • NVIDIA Jetson Nano
    • NVIDIA Jetson Orin Nano
    • NVIDIA Jetson TX2
    • Raspberry Pi (3+, 4 Model B)
    • Google Coral Dev board

  • Android Oreo or Newer

  • Others (Please contact support@cochl.ai)

If the platform you want to use is not listed above, please feel free to contact us at support@cochl.ai. We provide technical support to help you use our SDK on other platforms.

(2) Single-board computer setup

Set up the board with the NVIDIA’s official Getting Started guide.

Set up the board with the Google’s official Get Started guide.

(1) Download OS image

  • Download Ubuntu 20.04 64-bit at Ubuntu Downloads
  • (Raspbian can be another option, but make sure to use 32-bit version SDK)

(2) Flash OS image

  • Download and install balenaEtcher
  • Using Balena Etcher, flash image file into to microSD

(3) Run Raspberry Pi

  • Mount the microSD into the Raspberry Pi
  • Connect power, HDMI, and other peripherals (mouse, keyboard, ethernet …)
  • Network settings

2. Continue setup

Once your environment is ready, three steps remain:

  1. Install the SDK — download the archive from Resources for your target platform.
  2. Get your project key — create a project on the Dashboard and copy the key. See Dashboard Quick Start for the walkthrough.
  3. Launch the examples — follow the runnable samples for your language at Samples.

3. What’s next