The world's first fully open-source AI platform for ear-based sensing applications.
New OpenEarable 2.0.1 hardware available now!
Components
Tools, Libraries & Apps
Resources
Docs

True wireless audio.

 Biosensing.

 Tiny form factor.

OpenEarable is the world's first fully open-source AI platform for ear-based sensing applications with true wireless audio. Packed with an unprecedented array of high-precision sensors, OpenEarable redefines what's possible in wearable tech. Designed for both development and research applications, OpenEarable is modular, reconfigurable, and built for the future.

  Book Live Demo Free • No commitment
Full video

Engineered with precision.

USB-C Connector Coin Cell Battery 9-Axis IMU Air-Conduction Mic. 1 Push Button + RGB LED Pressure Sensor Bone-Conduction Microphone Air-Conduction Mic. 2 Dynamic Driver Speaker 3 Wavelengths PPG Optical Temperature Sensor
9-Axis Inertial Measurement Unit

Optimized for performance.

OPEN
EXTENSIBLE
EASY‑TO‑USE

TRUSTED BY COMPANIES AND RESEARCH LABS WORLDWIDE

… and many more.

Latest Research

Check out the papers below for the latest research published with OpenEarable. Want your project or paper to be featured here? Just send us a message!

CHOMP paper preview
CHOMP: Multimodal Chewing Side Detection with Earphones
J. Hummel, M. Burzer, et al.; Jan. 2026
EarResp paper preview
EarResp: Audio-Based On-Device Respiration Rate Estimation on Earphones with Adaptive Noise Suppression
M. Kuettner et al.; Jan. 2026
Paper preview
HearForce: Force Estimation for Manual Toothbrushing with Earables
Qiang Yang et al.; Dec. 2025

Which paper will you write?

Click here to find more papers about OpenEarable.

One platform

Limitless opportunities

True Wireless Earphones (via LEAudio)
Infra- to Ultrasound Sensing
Cardiac Activity (HR, HRV, SpO2)
Optical Skin Temperature Sensing
Body Sound (Bone-Con. & In-Ear Audio)
Body, Ear Canal, & Head Motion
Ear Canal Deformations via Pressure
edge-ml.org no-code machine learning
Which applications will YOU develop?

The most sensors in earphones. EVER!

SENSORS

POTENTIAL TO DETECT 30+ PHENOMENA
2 × Infra-/Audible-/Ultrasound Microphones (In Ear, Out of Ear)
Knowles SPH0641LU4H-1
OpenEarable features two microphones capable of capturing both infrasound and ultrasound audio signals. One microphone is oriented inward toward the ear canal, while the other faces outward. By default, it supports audio streaming to a connected device at up to 48 kHz and recording to the internal microSD card at up to 62.5 kHz.
Potential Applications
3-Axis Bone Conduction Accelerometer
Bosch BMA580
OpenEarable includes a 3-axis accelerometer that also functions as a bone-conduction microphone. Positioned at the entrance of the ear canal, it can also capture ear canal deformations. Recording is supported at up to 3.2 kHz for audio signals, making it ideal for detecting voice signals transmitted via bone conduction.
Potential Applications
Photoplethysmography Sensor
Analog Devices MAXM86161
The earpiece features a pulse oximeter with red, infrared, and green PPG sensors to detect blood volume changes via light absorbance. This enables real-time heart rate and blood oxygen (SpO2) monitoring. By default, OpenEarable supports streaming PPG data at up to 400 Hz via BLE and recording at up to 800 Hz. The calculated metrics can be streamed and recorded at up to 10 Hz.
Potential Applications
9-Axis Inertial Measurement Unit
Bosch BMX160
The 9-axis inertial measurement unit consists of an accelerometer, a gyroscope, and a magnetometer. Per default, OpenEarable allows streaming IMU data in real-time at up to 120 Hz via BLE and recording to the internal SD card at up to 800 Hz.
Potential Applications
Optical Skin Temperature Sensor
Melexis MLX90632
OpenEarable is equipped with a factory calibrated, medical grade optical temperature sensor that can measure skin temperature with an high accuracy. The data can be streamed via BLE or recorded to the internal SD card at up to 64 Hz.
Potential Applications
Ear Canal Pressure Sensor
Bosch BMP388
OpenEarable has an inward facing pressure sensor that makes it possible to measure pressure changes in the sealed ear canal. It possible to get precise measuremnts up to 120 Hz streamed via BLE and up to 200 Hz recorded to the internal SD card.
Potential Applications

HCI Interfaces

Dynamic Driver Speaker
Huayunxin RE8
The dynamic driver speaker inside OpenEarable supports playing audio via LE Audio from a connected device
RGB LED
Würth Elektronik LED0404
OpenEarable has an RGB color LED under the button. The color can be freely configured via BLE.
Button
TS2819A 230GF
The button on OpenEarable can be used to trigger events that are notified via BLE.

More Cores. More Power. More Possibilities.

Nordic nRF5340 - APPLICATION AND NETWORK

Dual-Core Arm Cortex-M33

The nRF5340 is the core processing unit inside OpenEarable, providing a powerful and flexible platform for ear-based sensing. This dual-core Bluetooth SoC integrates two Arm Cortex-M33 processors: A high-performance core (up to 128 MHz) is designed for control flow and data processing. A low-power core (up to 64 MHz) is optimized for background tasks and communication.
Analog Devices ADAU1860 - AUDIO PROCESSING

FastDSP Core & Tensilica HiFi 3z DSP Core

The ADAU1860 is a low-power digital signal processor (DSP) and audio codec used in OpenEarable, playing a crucial role in handling audio data efficiently. Manufactured by Analog Devices, the ADAU1860 integrates two DSP cores for real-time audio processing and high-fidelity signal acquisition. It creates the foundation for the development of future audio applications.

Designed for Expansion. Built for Innovation.

ARCHITECTURE

The OpenEarable device architecture is designed with a highly modular structure, enabling seamless integration of various sensors and components to adapt to different applications. The system's flexible interfaces and expandable design ensure easy customization, supporting diverse use cases.

DEBUGGING AND EXTENSION

OpenEarable is designed with modularity and expandability in mind, allowing researchers, developers, and enthusiasts to push the boundaries of ear-based computing. Its open hardware and software architecture enables debugging, sensor integration, and custom extensions.
  • Debugging: Using a breakout board, the 14-pin header connector on OpenEarable can be placed onto a standard 2.54 mm header connector. This allows for programming and debugging of the processing and DSP cores of OpenEarable.
  • Expansion: The 14-pin header connector offers 1.8V, 3.3V, GPIO, I2C, PDM, SoftSerial, SoftSPI, and SWD. The 12-pin FPC connector (BM46B-12DS) exposes 1.8V, 3.3V, GND, GPIO, I2C, PDM, Speaker_OUT, SoftSerial, and SoftSPI.
  • Storage: On the bottom side, next to the USB-C port, a microSD card slot is available for recording sensor data locally.

Open-Source. Free to Use. Always.

TOOLS, LIBRARIES & APPS

Firmware
OpenEarable runs on an open-source firmware based on ZephyrOS, a real-time operating system (RTOS) with modular architecture and real-time capabilities.
Mobile App
A mobile Flutter app for Android and iOS let's you control OpenEarable. It allows you configure OpenEarable and also offers a series of earable apps (e.g., a posture tracker).
Flutter Library
To develop custom tooling, the OpenEarable Flutter library provides all abstractions and simplifies iOS and Android development. This library is used by the official app.
Dashboard
The OpenWearable app is available as web deployment version. For the best experience we still recommend using the mobile Android app.
Apps
OpenEarable is not just a wearable device; it's a versatile platform that brings a myriad of applications. That's why we have developed example applications showcasing the potential of OpenEarable:
Apps
OpenEarable is not just a wearable device; it's a versatile platform that brings a myriad of applications. That's why we have developed example applications showcasing the potential of OpenEarable:
  • Recorder: Capture and analyze data from your OpenEarable device.
  • Posture Tracker: Get real-time feedback and improve your posture.
  • Heart Tracker: Basic example implementation to measure heart rate.
  • Jump Rope Counter*: Accurately count your rope skips during workouts.
  • Powernapping*: Monitor the quality of your rest with automatic sleep detection.
*Note: These apps were developed for OpenEarable 1.3 and are not supported by 2.0 at this time.

RESOURCES

OpenEarable is available as a fully-assembled third-party developer starter kit that includes everything you need to get started. You can also use our open-source design files.
Developer Starter Kit
Developer Starter Kit
The OpenEarable 2.0.1 starter kit includes everything that you need to get started with developing ear-based sensing applications. The OpenEarable starter kit includes:
  • OpenEarable (one pair, left and right)
  • 18 × silicone eartips (6 × S, 6 × M, 6 × L)
  • 2 × wing tip
  • 1 × dual-end USB-C charging cable
  • 1 × microSD card reader
  • 1 × travel case (small white box)
  • 1 × kit case (large black box)
  • 1 × programming adapter (2.54mm Dupon)
  • 1 × JTAG (2x10, 2.54mm) to SWD (2x5, 1.27mm) adapter
  • 10 × jumper wire (DuPont) female to female
Design Files
OpenEarable is completely open-source. You can freely change and adapt it to your needs.

DEVICE COMPARISON

OpenEarable 2.0.X OpenEarable 1.3 AirPods Pro 3
Audio
 Playback LE Audio via SD card BT Audio
Inward facing microphone 1 1 1 (no access)
Outward facing microphone 1 2
Bone Conduction microphone 1 1 (no access)
Sensors
 IMU 9-Axis + 3-Axis 9-Axis 6-Axis
PPG raw access to green, red,
infrared, ambient channels		 Heart Rate
Temperature skin + ambient ambient
Ear-Canal pressure
Memory
 Storage microSD card slot
+ internal SPI Flash		 microSD card slot
Battery
 Battery life up to 8 hours up to 4 hours up to 8 hours
Charging port USB-C microUSB charging case
Open-Source
 Hardware
Firmware
Software

DOCUMENTATION

Cheat Sheet

The OpenEarable cheat sheet contains all the basic information you need to get started quickly.



Data Processing

You can either record your data via the OpenWearable app as CSV files, or on the on-board microSD card slot in the *.oe file format ( download sample file). For *.oe files, use the Open Wearable Python library to parse recordings directly into analysis-ready data frames: 
pip install open-wearables
Quick start with SensorDataset: 
from open_wearables import SensorDataset

dataset = SensorDataset("recording.oe")
df = dataset.get_dataframe()
imu_df = dataset.imu.df
audio_df = dataset.get_audio_dataframe()



Programming

Comprehensive documentation on how to program the OpenEarable firmware is availabe on GitHub.


Frequently Asked Questions

Which smartphones support LEAudio?
Wireless audio streaming with OpenEarable requires an Android smartphone with LEAudio support (e.g. Google Pixel, >= Pixel 9 recommended). iOS currently does not support LEAudio.

Is it necessary to program OpenEarable before I can use it?
If you purchase the starter kit, OpenEarable comes pre-flashed with the latest firmware so you can immediatly use it with the dashboard, app and libraries. On a new OpenEarable, you will have to first flash the bootloader and firmware using a J-Link debugger. The firmware can be updated via J-Link and the 14-pin header connector.

How long does the battery last and how long does it charge?
The exact battery life will depend on your usage patterns, such as the frequency of data streaming and audio playback. Charging the battery to full capacity takes about 45 mins.

Are there any IP ratings for OpenEarable (e.g., water or dust resistance)?
OpenEarable does not have an IP rating. It's recommended to avoid exposure to water and dust to prevent damage to the electronic components.

What is the range of BLE connectivity for OpenEarable?
The BLE range for OpenEarable is consistent with standard BLE class 2 devices, which is typically up to 10 meters (33 feet) without obstructions.

How customizable is the firmware, and can it be updated wirelessly?
Our firmware is highly customizable and open source (Github repository). Firmware updates can be conducted through a wired connection. OpenEarable also supports wireless FOTA updates (experimental feature).

I am having issues with connecting my OpenEarable. What can I do?
Download a BLE Scanner for iOS or Android (e.g., nRF Connect for Mobile) and make sure you can find your OpenEarable. If you can find it, there is an issue with the device that you are using to connect to OpenEarable. Are you using the latest Chrome version? Does the OpenEarable app have the required permissions? Is the Bluetooth driver of your device updated?

Why can I not find any recordings on my microSD card?
In order to be compatible with OpenEarable the microSD card needs to be formatted as exFAT. Also we recommend to use the MICRON MTSD064AMC8MS-1WT microSD card, because other microSD cards can product electrical artifacts in microphone recordings.

What is the frequency response of OpenEarable's microphone and dynamic driver speaker?
The frequency response of the microphone can be found in the datasheet on page 6 and 7. The dynamic driver speaker characterization is available here.

CITING

OpenEarable has been published as a scientific research paper. If you are using OpenEarable in your work, please cite it as follows.
@article{roddiger2025openearable,
      title = {OpenEarable 2.0: Open-Source Earphone Platform for Physiological Ear Sensing},
      author = {Röddiger, Tobias and Küttner, Michael and Lepold, Philipp and King, Tobias and Moschina, Dennis and Bagge, Oliver and Paradiso, Joseph A. and Clarke, Christopher and Beigl, Michael},
      year = 2025,
      journal = {Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies},
      volume = {9},
      number = {1},
      pages = {1--33},
      publisher={ACM New York, NY, USA}
 }

PREVIOUS AND OTHER VERSIONS

The resources of previous versions and adaptations of the OpenEarable are available below.

PRESS KIT

If you want to report about OpenEarable, you can use the material from our press kit. Please make sure to credit OpenEarable and the "Karlsruhe Institute of Technology" as the source for all materials (CC BY-ND).
OpenWearables