RoboyImitator

Description

In this project we record emotions and voice from your personal computer and project those onto Roboy.

For speech-to-text, text-to-speech and emotion detection Microsoft Azure Cognitive Services are used. For speech-to-text speech_to_text/recognition_node.py creates a ros2 service at /roboy/cognition/speech/recognition.

Running RoboyImitator

This project consists of two parts:

1. Servers on your personal computer for emotion detection and microphone recording
2. Recognition node and main loop which connects to your servers and performs pyroboy.listen(), pyroboy.say("bla"), pyroboy.show_emotion().

Local installation

1. Run pip install -r requirements && pip install -e . to install the necessary python packages.
2. Create a file called secrets.yaml inside of the configs folder with the following keys and their corresponding values: stt_key, tts_key, face_key, service_region where the keys are from Microsoft Cognitive Services.

Roboy installation

docker-compose up --detach imitator