Since voice-based programs are slowly becoming the main way in which users interact with their gadgets now and later on, Ok Google’s voice search is already one step forward in the fight between its advanced features and those of its competitors’ virtual assistants, such as Microsoft’s Cortana and Apple’s Siri.
While the search engine’s app has traditionally obtained better results at knowing its users’ needs and responding to their questions perfectly, the organization now says that its application has now become even better at truly understanding the significance of the users’ concerns, as well. This means that the program has been enhanced in order to better understand human terminology and concerns that are more complicated.
Google first released the voice search feature in 2008, later incorporating it into the Knowledge Graph 3 years ago, helping it to answer all customers’ questions with actual details about people, location and other facts. Initially, it could offer basic data about public figures and then it was capable to respond to questions about their age or other related information, as Google describes its development.
Its capability to answer to various concerns was improved once again after some time, as it was discovered how to understand terms in different situations. For example, if users asked what components are in a certain tool, the virtual assistant realized what type of tool was mentioned in the respective context.
Now the search engines says that the voice program integrated into its proprietary app can comprehend the purpose of a question, so people will ask more complicated queries compared to previous versions. For this purpose, the company is splitting the phrase into different parts to better analyze the semantics in each one of these shorter questions.
Though any individual could quickly understand a simple query, to make a device able to perform the same operations is a challenging task. For example, the program has to understand the structure of a sports team, determine when that group of people won a specific trophy, then look into the season of that competition to choose the best response to the question in case.
The complex combination of elements is only one of the search engine’s new capabilities, the organization states. It can respond to difficult questions about a certain period, as well as queries about superlatives of any type. In the first situation, it means that users can ask the assistant questions with dates involved and it can answer a lot better than previous models. In the second case, the search engine now is aware of superlatives like largest, biggest and other requested terms.
If users tried to ask queries like these in the past, Google could have provided only results similar to the search phrases, without effectively determining which things they were truly asking about, according to the company. Basically, now is the first time when the search engine becomes a real semantic program, one that comprehends various contexts, as people normally and simply do, what humans are truly meaning when they have a question about something.
Image source: Cultofandroid