Honeywell awarded US contract to develop real-time speech anonymisation technology
The US government has awarded Honeywell a multi-million-dollar contract to lead the development of real-time speech anonymisation technology designed to protect individual privacy and improve communication in critical sectors such as aviation.
The project, known as Real-time Anonymisation and Speech Protection (RASP), aims to convert spoken language into anonymised speech. As well as helping protect speakers from identity exploitation, it could improve the intelligibility of heavily accented speech – potentially enabling air traffic controllers to more clearly understand pilots who are non-native English speakers.
Honeywell’s work forms part of the Intelligence Advanced Research Projects Activity’s (IARPA) Anonymous Real-Time Speech (ARTS) programme. The initiative is intended to help mitigate privacy threats from devices with always-on microphones – such as smartphones, smart TVs, and virtual assistants – that could potentially record and analyse conversations.
“By anonymising speech, RASP is designed to help address the growing risk of exploitation and privacy threats,” Honeywell said in a statement. “It will aim to protect an individual’s identity, as well as attributes like dialect, gender, and age, along with dynamic traits such as fear, stress, and anger.”
The ARTS programme is initially focused on English, with future phases set to expand into Spanish and other languages. Honeywell is working alongside researchers from the University of Rochester, Texas A&M University, and the University of Texas at Dallas.
“With voice technology becoming commonplace in every home, office, and public space, our voices can reveal key information about our identity and emotions that may not be appropriate for others to know,” said Tor Finseth, principal investigator for RASP at Honeywell Aerospace Technologies.
“Until now, no one has been able to anonymise speech in real time while also making it easy to understand,” he added. “Through our RASP system, Honeywell aims to meet government requirements to protect against speaker identification tools, human listeners, and machine-learning assessments. We also aim to improve anonymous speech latency, understandability, and naturalness.”
The project underscores a growing focus on privacy protection as artificial intelligence and voice-recognition tools become increasingly embedded in daily life.
