Voice & Data Convergence

What must change before 'voice-first' hits the enterprise?

This is a contributed piece by Craig Walker, Director of Cloud Services at Alcatel-Lucent Enterprise

The likes of Siri, Cortana, Alexa and a host of other personal assistants are making a big noise in the consumer market. Whether it’s to play music, order shopping or tell some jokes, voice-controlled devices with their own assistants are bringing new and exciting ways to interact with technology. But there are also real business benefits of these systems as we learn to leverage voice technology to help simplify work and optimise productivity in a range of different industries.

For example, giving hotel guests the option to say “System: Turn off the lights in the bathroom” or “System: Let reception know that I’m about to check out” could greatly improve guest experience. A rail engineer working on an underground system could ask for the latest repair information to be brought up on his handheld device without having to down tools. A teacher in a classroom could tell the computer to bring up information on 18th century history, or tell the system to turn on the projector, close the blinds and play a video, all without having to leave their seat.


Voice technology – closer than we think?

So how close are we to really seeing these examples? For a voice-first environment to fully enter the enterprise, a few things will need to happen first.

How does voice search need to evolve to reach mass adoption? The rise of voice search: What do businesses really need to know?

Security and privacy are top priorities for businesses. Proving these can be dealt with compliantly will be critical if we are to start having our enterprise systems relying on voice commands. Should anyone be able to command critical equipment or systems just by speaking? The answer, clearly, is no. We therefore need to start thinking about the regulations surrounding the use of voice technology. Are a person’s rights violated if these verbal commands expose their information to third parties?

Some devices, such as smart home systems, will be always-on, always listening and potentially recording. A few well publicised cases of privacy invasion, commercial espionage or legal jeopardy could stall adoption. This suggests that a big on/off switch or function needs to be included in voice-first products, so that users may get the benefits without risking the downsides of constant monitoring. Secure software access would also need to be built into solutions to prevent and detect hacking efforts.


Context is key

The most popular use cases of voice tech revolve around response systems such as the ones found in cars and smartphones. But many of us know that these systems work marginally at best, especially if there is background noise, different languages or accents. Recognition and contextualisation need to be refined through technological developments before we can realistically think about enterprise-wide adoption.

Microsoft recently announced that its speech recognition system had reached the lowest ever failure rate of just 5.1 percent. But with over 170,000 words in the English language, that’s still nowhere near extensive or accurate enough, especially when considering the delicate operations voice systems could be used for, such as in hospitals or banks.

The problem isn’t just voice recognition though, it’s also about what to do with the words. This is where artificial intelligence (AI) comes into play - helping understand and bring context to the words.

AI can be used as a check and validation system, for example a doctor who said to prescribe 200 grams instead of 200 milligrams or an engineer telling a system to turn off a cooling system to a nuclear reactor. These actions initiated by voice systems might have life-threatening consequences, but it won’t know unless it has contextualization and broader intelligence to stop that from happening.


Voice security

Despite these challenges, we are already seeing voice systems able to support secure access from authorised users. Banks in the UK started to introduce the technology last year, recognising a customer’s voice and formation of words to negate the need for security questions and passwords.

I expect we will continue to see improvements in voice recognition systems in other industries that will enable voice security to be viable in an enterprise environment and ensure that only authorised users can perform the associated actions.


Speaking up for CPaaS

In the enterprise world, we are seeing the explosion of Communication Platform as-a-Service (CPaaS), leveraging APIs to transform today’s applications into voice-integrated solutions. Some CPaaS platforms now offer a standardised set of APIs to enable companies to integrate communications into their business processes.

Integrating voice and video services into existing applications will play a big part in that “voice-first” enterprise environment by leveraging the rich API infrastructure of CPaaS to communicate to applications and things.

But these CPaaS platforms will need to be standardised before we will see rapid development of voice technology. Each of today’s consumer-based voice-controlled systems have their own interfaces and API integrations and, as with many of the historic technology battles of the past – floppy disk vs. USB stick, DVD vs. streaming services - this may lead to product obsolescence. Just as a consumer doesn’t want to buy a ‘smart fridge’ only to find the technology that controls it was just discontinued, an enterprise wants to ensure that their technology investments won’t be obsolete before they are able to realise a return.

There are already a set of technologies in the works to help minimise this potential obsolescence. Frameworks such as IoTivity are being developed to build a standardised platform, and we are already seeing the value, benefits and rapid expansion of new voice applications for businesses.


The best is yet to come

In the future, I am certain that we will see some of the basic use cases move into the enterprise. As advances continue to be made in voice recognition, voice security and standardisation in device connectivity, we will see more and more voice-first activities in the enterprise to help reduce complexity and improve productivity.


« News roundup: Why this week was bad Krack for Wi-Fi


Blockchain for insurance: A realistic picture »
IDG Connect

IDG Connect tackles the tech stories that matter to you

  • Mail


Do you think your smartphone is making you a workaholic?