The Australian Army and Microsoft are proud to announce they are collaborating on a program that explores how modern digital technologies could transform command and control in the future. The initial focus is on using Artificial Intelligence (AI) to transcribe and analyze voice communications over combat net radio. Automating battlespace communications transcription will also free personnel for more human-centric tasks. The Command, Control, Communications Process Optimisation initiative aims to bring clarity to what can be a chaotic operational environment – with reports and searchable transcripts providing commanders with improved situational awareness. The resulting insights will help optimize operational decision-making and create important and intelligent digital foundations for the transformation of AI-enabled decision-making over the next decade.
Brigadier Ian Langford, Director General of the Future Land Warfare Branch said; “Employing artificial intelligence tools through a reliable, secure and resilient platform to consolidate battlefield communications is an important step toward the creation of a decision support engine to create decision advantage in the future. Collaborative and innovative methods to achieve this accelerate the rate of adoption of cutting-edge technology to afford an Army in Motion an edge in the future.”
This is highlighted in the Army’s Robotic and Autonomous Systems Strategy (RAS Strategy) under the Improving Decision-Making banner. The team, comprising of Microsoft system analysts, data scientists, and data architects along with the Army’s Robotic & Autonomous Systems Implementation & Coordination Office (RICO) personnel from the Future Land Warfare Branch, is presently prototyping a solution to transcribe combat radio communications to text, identify key message parameters such as the sender and recipient and to extract meaning from the transcription. Artificial intelligence will be used to interpret and search communications to infer insights about the battle with an end-user presentation layer through Power BI.
The prototype solution deploys Azure cognitive services to support transcription and natural language processing. Due to the nuances of radiotelephony communications, significant customization has been required to ingest, clean, and filter audio signals. The prototype has been built in the Army Land Network Integration Centre (LNIC) Azure environment and is the first example of this platform being used in a collaboration with industry. Designing, testing, and validation have been performed with HQ 3 Brigade as the ‘end user’ for the program.