Voice assistants: a different platform
Spanish journalist Marta G. Aller called me to talk about voice assistants such as Amazon Echo or Google Home for an article in El Independiente (in Spanish).
Although not yet launched in Spain, use of these types of home assistants is growing strongly in the United States, and set to take off this year. They were launched in September 2016 in the United Kingdom and Germany.
To all intents and purposes, voice assistants are a new platform: their growth depends on attracting developers and applications able to incorporate new skills. One thing is using a voice assistant to make a shopping list, but they can also be used to choose songs on Spotify, turn household lights on and off, raise or lower the central heating or air conditioning, ask for a pizza, call Uber or read you an audiobook, among thousands of other possibilities.
It is also important to understand the important differences and limitations of a voice assistant compared to the screen, which we have become so used to: a voice assistant simply reads out one or two options, any more would be too complicated, compared to the multiple possibilities offered by the screen, which can lead to problems related to the selection of those options, as Google could recently verify. When requesting something from a voice assistant, we should remember that we will get fewer answers and choices than our usual screen interface would offer us.
Amazon didn’t create Echo just to help us with our shopping, as its use shows: people use it for a wide variety of tasks, from programming an alarm to reading the news, passing the time, putting on the lights or even telling a joke. It is precisely the speed of development of such applications and rapid growth in use that points to a speedy adoption process.
These are advanced voice processing systems based on machine learning that rely heavily on the user base and the level of use for launch in a new language or to improve their performance, hence Google’s advantage in having Google Now’s user base on Android-based smartphones (allowing it, for example, to include voice recognition for up to six users), or Amazon’s decision to use voice interface on its smartphone app.
As such , developing the algorithms for Spanish should not take long. From now on, we must think of these devices as the center of an ecosystem, as a platform within which to integrate a growing range of functionalities that will make voice assistants increasingly useful and more attractive, albeit not exempt from teething problems.
(En español, aquí)
