This technology isn’t new to us; we’ve been using voice-activated technologies for quite some time now. You may have also heard of voice recognition; which is actually used to describe both speaker and speech recognition. Speaker recognition is when the device can detect who is speaking and speech recognition is when the device can detect what is being said.
The act of activation means that whatever is being said, or whatever sound is being made, it is triggering a reaction. There are several examples of this within our everyday technology, from our mobile phones to games consoles and web browsers. Let’s delve deeper and discover how this technology is being used in the present day.
Have you ever called a cinema to book tickets and were met with a robot asking you to tell them what movie you’d like to book? This is speech recognition at its finest and it’s used by millions of companies and businesses worldwide to tighten up on customer service. Gone are the days where you talk to a human being, you can book or reserve almost anything without having to talk to a single person. Almost.
There are however several downsides to speech technology, firstly the way in which it detects your voice. We’ve all had one of those moments where we don’t quite know what someone has said, right? Well, the same happens here, the machine can only go so far and if you have a distinct accent it may also become more of a struggle to get anything done. So much more time is being invested in improving speaker recognition over speech recognition as it is more precise.
A good example of speaker recognition would be an announcement made in 2013 by Barclays Wealth, declaring that they were to use this technology to detect and verify the identity of their customers within only 30 seconds of conversation. This technology, created by Nuance, was rated very highly by Barclays customers who gave it on average a 9 out of 10. They claimed that is was easy to use, was super-fast and most importantly had fantastic security features.
Speaker recognition has also been built into the very latest smartphones, led by the mac daddy of mobile technology, Apple (using personal assistant Siri). With Siri, you can not only use your own voice but the voice of others with your phone, the speaker recognition used means that it can detect exactly who is talking. However, there is a new kid on the block in the form of Cortana, a personal assistant for newer Windows devices that was released to the public in October of this year.
Siri has a cool feature, which is that when charging you can talk to your phone (even when it is locked) by saying “Hey Siri” and asking a question or giving a demand (which is hit or miss). Cortana, however, has an even better feature called Cortana’s Notebook, which is where you can tell Cortana all about yourself and your interests, helping to make using ‘her’ more of a breeze
It’s only very recently with the latest games consoles from Playstation and Xbox that you will find voice recognition. This technology, however, doesn’t stretch to speaker recognition but uses the same type of speech recognition used to book over the phone. This means that they do not have the power to determine who is talking, which means that anyone can command your console.
You can also only use voice recognition if you have a microphone attached to the console, which can usually be found on a camera or a headset. The commands are basic and pre-set, you can say the name of a game or you can use media terms such as “start”, “stop” and “pause”. It’s good if you’re an avid gamer, but for the average Joe, it’s just another fancy feature.
This is where voice recognition takes a big leap forward, with the introduction of “Okay Google”, a new way to search online. Instead of typing in a search query, you can actually use a microphone to ask Google a question, which in turn turns into a Google search. This is great for people who are on the go and you can use this type of searching the web on Android devices, iPhones, iPads and any other device with a Chrome web browser.
The search isn’t limited to those who speak English either; you can also start a search in French, Italian, German, Russian, Korean, Japanese, Spanish or Portuguese. This technology also works better for those with a distinct or strong accent, resurrecting modern-day speech detection.