The Omda Incident business unit provides robust systems for managing every aspect of Emergency Response. The Incident product solution supports case management, GIS (geographical information and maps), dispatching of resources and communication between all parties involved in the emergency response chain. It is deployed in control rooms, e.g. for 112 organizations.

Every year, the Swedish national emergency number (112) handles 3.5 million calls. These time-critical calls are handled by operators, who must make split-second decisions while interpreting the help seeker’s description of events and recording the details into a Computer-Aided Dispatch (CAD) system. To save more lives. Every second, every day, every year.

Our solution is already able to transcribe calls in real-time. This thesis aims to build upon the transcription feature, to enable the system to recognize specific entities (e.g., names, addresses) in the call. The expected impact is to help operators become more efficient, less error prone and having more time for other calls.

The current solution is able to separate audio streams in a conference call and transcribe each audio stream. This means that the algorithms can be trained on each participant in isolation as well as the combined call.

Master thesis objective:

The objective of this thesis is to develop models for ML/AI solutions that can improve the speech-to-text feature to decipher spoken language in emergency situations, categorize key information in preparation for an integration in the existing CAD system.

The student(s) will select, investigate, and develop models for Natural Language Processing or, more specifically, the subarea of named entity recognition. Entities of interest include but are not limited to:

• Social security numbers and names
• Phone numbers
• Addresses and landmarks

Our tech stack:

• The model will be developed using suitable tools (e.g. Tensorflow) and languages (e.g. Python), to train, tune and validate the model.

What we offer:

• Access to data and development environment
• Mentor and domain knowledge support from Omda employees

Competence requirements:

• Data Science (AI/ML theories and models)
• Experience in programming in Python
• Experience with machine-learning and AI projects
• Self-going and problem-solving
• Language: proficient in both Swedish and English
• Swedish Citizenship

Looking forward to meeting you!