Voice assistants useful, but adoption slow

Voice assistants useful, but adoption slow

Summary: Personal voice assistants (PVAs) such as Apple's Siri could represent the next big thing in the tech industry, but lack of defined use case, and difficulty in collating all of the world's spoken languages into a database, mean widespread uptake of the tech is unlikely to be soon, observers have noted.


Personal voice assistants (PVAs) such as Apple's Siri could represent the next big thing in the tech industry, but lack of defined use case, and difficulty in collating all of the world's spoken languages into a database, mean widespread uptake of the tech is unlikely to be soon, observers have noted.

Vikas Chanani, industry analyst for ICT Asia Pacific at Frost & Sullivan, said that as a technology, the PVA is the "next computing paradigm shift" in the mobile-devices space.

"Introducing artificial intelligence [AI] to a smartphone in the form of the PVA is a big leap toward how we'll come to communicate with devices, and opens up a plethora of opportunities for all of us," he added in his email.

Basic capabilities, such as placing phone calls, sending SMSes or emails and setting reminders, are already available with existing PVAs, but these are only "the tip of the iceberg", the analyst noted.

He said that beyond mobile phones, the technology could also be integrated into computer games, automobiles, navigation devices, music players, TVs and home appliances. Chinese PC manufacturer Lenovo, for example, had launched its K91 Smart TV that comes equipped with voice-control features, earlier this month.

"Companies and developers are working to harness each bit of this technology, finding ways to integrate conversational voice interactions into products, applications and services [that will] enhance the capabilities of PVA going forward," Chanani stated.

"The personal voice assistant is no flash in the pan. We're talking about another technology revolution, which will have implications on several industries and even human society in the way we communicate and perform our day-to-day activities."

Tang Pin-Chen, research analyst at Canalys, concurred. He noted that PVAs could "change the way we work, how we use our devices and enhance the user experience" in the future.

That said, businesses will have to invest a significant amount of time and money in order to realise the potential, he added in his email.

Siri sets the tone

Both analysts pointed out that while voice assistance is not new to the field, Apple has given the technology a boost when it introduced Siri alongside its iPhone 4S in October last year.

Chanani said that with Siri tightly integrated with the Cupertino giant's iOS platform, it made the tool more user friendly, and helps it to recognise human, conversational speech commands, unlike other voice-recognition tools then.

Tang added that Siri has made interactive AI part of the user experience, rather than just having a device that obeys simple commands with limited response.

"With Apple championing this innovation ... interactive voice assistants will become a standard across mobile platforms in the near future, because competitors do not want to be left behind," he surmised.

Web titan and rival Google was reportedly working on its own PVA for the Android platform, codenamed Majel. Developers, too, have been populating the Android Market with Siri clones, according to an earlier report by PC World.

Mass adoption not likely soon

There are barriers to the widespread use of these voice assistance apps, though, said Tang. This is because such apps need a well-defined usage scenario, which is currently limited to just the few, such as navigation while driving, scheduling appointments and text messaging, he noted.

"Consumers are still not taking to the idea if using personal voice assistants, and prefer to do the input themselves, since it is faster and easier," the Canalys analyst stated.

PVAs are also dependent on the ambient quality of the surroundings, he added, saying that a poor setting would mean that the app would not clearly and accurately pick up speech commands. The vast number of accents and languages spoken in the world would mean that such apps would require a huge database of information in order to work universally, Tang stated.

"Personal voice-assistant apps do possess untapped innovation potential, but these limitations mean [the technology] will not be dominating headlines in the short term," he concluded.

Mobile app developers who spoke to ZDNet Asia also expressed similar reservations. Chua Ziyong, a Singapore-based Android developer and CEO of mobile-solutions company Stream Media, said that while Siri has given Apple a "huge edge", and PVA looks set to be the next big thing, this will not happen in the immediate future.

He said in an email that the technology is only just maturing, and standards will need to be created for integration and interoperability with other appliances.

"The question is always about reliability. Most voice products in the market are not reliable, and the high percentage of false positives and failures leads to consumer frustration and reduced adoption," Chua explained.

He also pointed out that users have to be "educated" on how to speak in specific tones, accents or words to get the product to work, which further limits the adoption and use case for PVAs.

Karen New, CEO of mobile app-development company Omnitoons, added that the accuracy of voice recognition has not yet reached 100 per cent, considering that accents and colloquial terms will need time to be indexed.

People would also need to get used to talking to machines more often, which most find "weird" in public places, she noted.

Hence, Chua said it may take "another two to three years before we see a reliable and intuitive use of this technology".

Via ZDNet Asia

Topics: Apple, Collaboration, Google, Unified Comms

Jamie Yap

About Jamie Yap

Jamie writes about technology, business and the most obvious intersection of the two that is software. Other variegated topics include--in one form or other--cloud, Web 2.0, apps, data, analytics, mobile, services, and the three Es: enterprises, executives and entrepreneurs. In a previous life, she was a writer covering a different but equally serious business called show business.

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • Doh!

    That they actively make the user interfere with those around them is a big enough mitigator against wide adoption.

    While many don't seem to know about anything outside of the loud conversation they are having on their mobile, most would feel intimidated interacting verbally in the open about private details. One cannot mumble a request to these AI programs - loud and clear is the requirement.

    How did Face Time work out for most users. It was a 'big thing' Apple heavily promoted. Trouble is, that unless you are satisfied with others seeing you 'warts-and-all' (less likely with age) and trying to make sure that you are keeping the camera on you all the time, the novelty wears off very quickly.

    I suspect that while it sounds 'cool', the probability of it being frequently used by many people is slim.
    • Mass adoption would probably bring a significant regulatory backlash as most would be asking many of the same basic inane questions.

      While it would be hard to legislate against similar parent-child conversations, such annoying people-machine interactions would be easy to class as noise polution.

      Open plan offices have already opened up the amount of interfering noise workers must encounter, but increased popularity of IV programs would require a return to noise isolating cubicles, which hikes office costs up and flexibility down, resulting in more businesses likely to ban or severely restrict them.
  • The problem with IR is that widespread popularity would exacerbate word recognition, especially as many nearby would be using the same core sets of words.

    Imagine a person who has a heavy accent being p!!ssed off when their phone responds better to a nearby more clearly enunciating speaker!
    • Meant IV, not IR.