Overview

This project provides a bridge between the Thalmic Myo Armband and OSC connected applications such as Pd, SuperCollider, or Max/MSP. The code is purely Python designed to run in MacOS and Linux using the Myo bluetooth dongle.

Please note that this project requires Python 3, and uses the pyserial and python-osc modules.

Usage

Install the dependencies:

pip install -r requirements.txt

Start the program:

python3 myo_to_osc.py

It will connect to the first Myo it sees and then start sending OSC messages for the EMG and IMU sensors.

Open rec_myo.pd in Pure Data to see an example of reading these OSC messages.

To quit, type Control-C in the Python session.

Other optional arguments such as specific connections and loggin are available, try python3 myo_to_osc.py -h for all options:

usage: myo_to_osc.py [-h] [-l] [-d] [-a ADDRESS]

Connects to a Myo, then sends EMG and IMU data as OSC messages to
localhost:3000.

optional arguments:
  -h, --help            show this help message and exit
  -l, --log             Save Myo data to a log file.
  -d, --discover        Search for available Myos and print their names and
                        MAC addresses.
  -a ADDRESS, --address ADDRESS
                        A Myo MAC address to connect to, in format
                        "XX:XX:XX:XX:XX:XX".

Dongle device name

This program requires the Myo's included USB Bluetooth LE dongle which provides a simple serial interface for connecting to the Myo.

The BT class should be able to find the serial port of the dongle automatically, if it doesn't try unplugging and plugging it back in or specifying the address manually. You can start the Myo class with the adapter's device name as an argument if you want to be sure.

Pose / Arm Classification

This library can use the Myo's onboard pose and arm recognition. You have to turn this feature on using the set_mode function of the Myo class. e.g.,

m.set_mode(EMG_Mode.send_emg.value, IMU_Mode.send_data.value, Classifier_Mode.enabled.value)

Then perform the sync gesture as described by Myo Support:

Make sure you're wearing Myo with the USB port facing your wrist. Gently flex your wrist away from your body. Myo will begin to vibrate when it recognizes this gesture. Hold this gesture for a few seconds until Myo stops vibrating.

You will know you performed the sync gesture successfully when the Thalmic Labs logo LED on the armband stops pulsing. If it needs to warm up, you will see it blink along with an notification next to the gesture indicator window in Myo Connect. Once Myo is fully warmed up and synced, you will feel three distinct vibrations.

About the Myo Bluetooth dongle

The Myo Bluetooth dongle provides a simple serial interface to Bluetooth LE devices, thus the code here can be relatively simple and cross-platform. There are Python libraries for native OS-specific Bluetooth LE connections, but I haven't found one that works better or as well as the present solution.

There's a working version of this repo using the pyGatt library that can also use the dongle on any platform and other Bluetooth adapters under Linux (see the pygatt-version branch), but I found the performance to be too bad to use.

Another possibility would be to use the Adafruit Python BluetoothLE library, which could work under Linux and MacOS but not Windows.

TL;DR: This library requires the Myo Bluetooth dongle for good reasons.

Citation

If this project is useful for you in academic work, it would be wonderful if you could cite the paper where it is introduced:

Charles P. Martin, Alexander Refsum Jensenius, and Jim Torresen. Composing an Ensemble Standstill Work for Myo and Bela. In Proceedings of the International Conference on New Interfaces for Musical Expression, NIME '18, June 2018.

@inproceedings{Martin:2018ab,
    title = {Composing an Ensemble Standstill Work for Myo and Bela},
    author = {Charles P. Martin and Alexander Refsum Jensenius and Jim Torresen},
    booktitle = {Proceedings of the International Conference on New Interfaces for Musical Expression},
    year = {2018},
    series = {NIME '18},
    url = {http://folk.uio.no/charlepm/preprints/2018-ComposingEnsembleStandstillWork.pdf}
}

Acknowledgements

Thanks to the original authors of the myo_raw library, and later contributions that served as a starting point for this project: Danny Zhu, Alvaro Villoslada, Fernando Cosentino.

License

This project is licensed under the MIT License.