To get the things done with the Digital Assistant of Google, it needs to understand you, it has the duty to identify the words that you are saying and it also has to understand what you mean by saying that context. It is very hard to understand the language especially when the speaking style varies from person to person. It irritates the person when he tries to send a text or make a call and the assistant of Google mispronounces or simply does not identifies the context. Now, Google has made improvements to its assistant that can help this digital assistant to better understand unique names different contexts in the conversations shared by users. The search giant will apply its (BERT) system to advance the quality of the conversation between Google Assistant and the person involved in the conversation with the assistant.
This improved version is currently available on the Google smart speaker in the US and this will be rolled out to smartphones and smart displays in the coming future. The basic aim of this improvement is Google wants its assistant to be user-friendly as it tries to better recognize the speech and pronounce the unique names than it did earlier. Google has also said that users can now teach the digital assistant to pronounce and recognize unique names as the company says further, the assistant has the ability to understand what the users are saying and it will remember that pronunciation in the future without the recording of the voice of the user. However, this improved feature is still available in the English language only but the company is planning to roll out this feature in other languages in the future.
Another feature with is also improved with this update is the ability to recognize the speech of the user and the context of the dialogue in an informal manner. For this Google has rebuilt Assistant’s Natural Language Understanding (NLU) models so it can comprehend context while also refining its ‘’reference solution’’. Google says that this update uses the machine learning technology powered by BERT and due to this, the company claims that its assistant can respond with almost 100% precision to the instructions for alarms and timers. Google is now working to expand this feature to more devices than just Google’s smart speaker that is recently available. The company also applied BERT to make the conversation even better than before between the assistant and the user, the assistant will now use the previous interactions of the user and it will recognize what is being shown on the screen to respond to further questions supporting in more natural conversation.
Read next: Google Report Reveals Significant Year on Year Increase in Search Spam Detection
This improved version is currently available on the Google smart speaker in the US and this will be rolled out to smartphones and smart displays in the coming future. The basic aim of this improvement is Google wants its assistant to be user-friendly as it tries to better recognize the speech and pronounce the unique names than it did earlier. Google has also said that users can now teach the digital assistant to pronounce and recognize unique names as the company says further, the assistant has the ability to understand what the users are saying and it will remember that pronunciation in the future without the recording of the voice of the user. However, this improved feature is still available in the English language only but the company is planning to roll out this feature in other languages in the future.
Another feature with is also improved with this update is the ability to recognize the speech of the user and the context of the dialogue in an informal manner. For this Google has rebuilt Assistant’s Natural Language Understanding (NLU) models so it can comprehend context while also refining its ‘’reference solution’’. Google says that this update uses the machine learning technology powered by BERT and due to this, the company claims that its assistant can respond with almost 100% precision to the instructions for alarms and timers. Google is now working to expand this feature to more devices than just Google’s smart speaker that is recently available. The company also applied BERT to make the conversation even better than before between the assistant and the user, the assistant will now use the previous interactions of the user and it will recognize what is being shown on the screen to respond to further questions supporting in more natural conversation.
Read next: Google Report Reveals Significant Year on Year Increase in Search Spam Detection
