Commands and call placement are done through voice commands issued to the Vocera badge (e.g., Call Frank Bulk). Vocera licenses voice-recognition technology from Nuance Communications. The voice recognition adds to the coolness factor of the devices, but also represents an Achilles' heel. There's no question that voice response has improved significantly in recent years, and Vocera has done a credible job of exploiting its power. However, since the voice commands themselves are digitized and sent over the wireless network to the server's speech-recognition engine (an 8-Kbps codec is used during conversation, but a full 64 Kbps is used to get the best quality when using voice commands), high packet loss and high background noise can cause the system to have trouble processing voice commands. Using a headset with a high-quality microphone can help. Under normal conditions, voice recognition proved extremely effective, but during some of our harsher simulated network conditions, the badges became a little difficult to use as the quality of voice recognition dropped.
Vocera relies on WEP for encryption services together with a sophisticated optional voiceprint-based authentication system. Once you initialize the system, it compares subsequent login attempts to the stored voiceprint. Vocera claims the system will continue to refine the voiceprint through subsequent use and eventually "capture the behavioral characteristics of the way a person speaks." This is a relatively new form of biometrics and our first opportunity to evaluate such a system. Aside from a slight delay during authentication, the system worked as advertised during testing. Vocera's Web site includes a white paper that provides detailed implementation guidelines designed to ensure HIPAA compliance, which is important given the company's target health-care market.
We easily integrated the Vocera system into our Centrex environment by using the telephony server together with a Dialogic analog voice interface. Using voice commands, we were able to request outside lines and make phone calls on the PSTN. It all worked without a hitch.
Overall, call quality with the Vocera badges was very good, despite the system's low-bandwidth requirements, particularly considering the microphone is usually located somewhere around the user's chest. Even under simulated network congestion, call quality remained consistently good. Range was also impressive, particularly considering the small size of the Vocera badges. Layer 2 roaming delays were virtually imperceptible.
Vocera's VoIP solution represents one of the more extensible solutions we tested. The broadcast message functions, user location and voice-recognition approach are some of the most innovative ideas we've seen. The only drawback to the Vocera approach is that voice recognition requires a good amount of both server and network overhead because the commands are issued over their own data streams. For those who find standards important to their enterprise, Vocera does not currently use SIP, but plans to add limited support for it in the future.
To realize the benefits of cloud-native software, an application must actually be cloud-native—designed to run in the cloud, interacting with disaggregated cloud services, fully manageable as code—to deliver the expected benefits.
By combining resources and expertise, the AI-Enabled ICT Workforce Consortium will offer a blueprint for how industries can adapt to an AI-dominated future.
IT leaders should heed the guidance of cybersecurity insurance providers who think businesses should prioritize security education, incident preparedness, regular internal audits, and ongoing vulnerability scanning and patching.
