Short Description

At the SPSC, we have built an voice controlled home automation system for Austrian German in our meeting room and kitchen. Now, we want to use this as a testbed for students to try own related work in a real environment. Unfortunately, the current system evolved from a ASR simulation system and is not easy to modify. Therefore, we want to redesign the system and make it modular to be able to include and change components, such as signal pre-processing (beamformer, speaker lokalisation, …), automatic speech recognition, and a dialog manager.

Your Tasks

• Review of the current system
• Design framework, interfaces for the system
• Implement the core system
• Documentation

Your Requirements

• Motivation and interest in the topic
• Speech communication background
• Good Matlab (or Python) skills
• Recommended: Real-time audio processing background
• Recommended: Software development affinity

Idealy the project team should consist of at least 3 students.