Military Forces find themselves operating in an increasingly complex environment: personnel must efficiently manage and react to multiple streams of information. Whether by conveying relevant information on the battlefield or by training in simulation exercises, efficient vocal interfaces are key to providing highly sought human-technology synergy.
During Operational missions, military personnel must manage multiple channels of information: outputs from a many different sensors and systems, communication with fellow troops, orders from commanding officers. This informational barrage may be overwhelming and make it difficult to ascertain precisely what information is vital at any given instant in time. Moreover, the inclusion of visual displays is often counterintuitively inefficient in such contexts as the attention and cognitive load on the user are heavily solicited. In contrast, Voice offers an efficient and robust access to key information. Delivery of the most useful messages is prioritised, minimising effort on the part of the recipient.
Soldiers and commanding officers need frequent updates on their environment: location of potential threats, positions of fellow troops etc. A voice-enabled interface can convey such information in real-time – the message is received as soon as it is available, not once it hap- pens to be read. Furthermore, any flow of information can be dynamically rescheduled with higher-priority messages able to interrupt those less urgent.
Voice Synthesis is easily available on multiple channels: either centralised on a server to convey the information to the desired personnel, or embedded on a specific device (a tablet, a helmet, smart speakers, etc). Having the same voices on all channels helps you to preserve the same vocal experience on all your Man-Machine Interaction capabilities.
In operational conditions, there is no room for ambiguity. Voxygen’s speech synthesis maximises the intelligibility of messages in two different ways: our high-quality voices are designed with clear articulation in mind and we can further reinforce the intelligibility of a specific voice with a speech enhancement post-processing module.
But intelligibility alone is not enough. Context matters. The tone and expressiveness of a synthetic voice must be flexible so that any message’s intent may be clearly understood. Our exemplary voice synthesis can cover a broad range of speaking styles to suit the situation. Where discretion is critical, our voice can comply by whispering. Orders can be delivered in an authoritative tone, emergencies can be reflected with a more alerting style, congratulations can be given with warmth. Adapting the expressiveness to context facilitates a more efficient, pertinent and unambiguous man-machine interaction.
When combined with speech recognition and translation technologies, speech synthesis enables two individuals who do not share the same language to communicate with ease. Such speech-to-speech systems are of particular interest when military forces need to exchange with local populations, e.g. at check points or in case of emer- gency conditions (pandemic alerts, natural disasters, etc). A speech-to-speech system can be easily deployed on a tablet with the desired languages.
