WorkingAge voice service

Codebase for the voice service of the WorkingAge tool

Speech emotion recognition

This is the code of the module for speech emotion recognition integrated in the PoliMI edge server. The module communicates through the ZeroMQ platform. The generated JSON payload with the recognized emotion has the following format:

{
  "probability": 0.79340570,
  "timeStamp": "20211231235959",
  "sensorType": "Microphone",
  "values": 
  {
    "sensor": "EmoState",
    "value": "Positive"
  }
}

The payload is sent through the ZeroMQ publisher using the user pseudo ID as topic, e.g. U550e8400-e29b-41d4-a716-446655440000.

Installation

We suggest to use Python 3.7 as environment. FFmpeg is expected to be installed and present in PATH

To install the required packages use the requirements.txt file.

pip3 install -r requirements.txt

If you're not using a virtual environment, it is suggested to install with the user-only option

pip3 install --user -r requirements.txt

Usage

Service can be run in different modes.

Background

To run in background mode

nohup python3 workingAgeVoiceService.py &

SystemD

To run as a SystemD service (preferred)

sudo systemctl enable workingage.service
sudo systemctl start workingage.service

Service configuration file should be /lib/systemd/system/workingage.service for Debian/Ubuntu-based systems and in /usr/lib/systemd/system/workingage.service for CentOS/Fedora/RHEL-based systems.

Additional commands to interact with the SystemD service

journalctl -u workingage  # check status

sudo systemctl enable workingage.service
sudo systemctl disable workingage.service
sudo systemctl start workingage.service
sudo systemctl stop workingage.service
sudo systemctl restart workingage.service
sudo systemctl status workingage.service

For starting and configuring the workingAgeVoiceService.py as a systemd service, see:

• https://tecadmin.net/setup-autorun-python-script-using-systemd/
• https://www.golinuxcloud.com/run-systemd-service-specific-user-group-linux/

Foreground

To run in foreground mode

python3 workingAgeVoiceService.py

Additional notes

During tests, when using a local ZeroMQ proxy, to check its status issue the command:

netstat -a | grep -e:5559 -e:5560