Google working on new technology to improve voice assistant feature

 Google working on new technology to improve voice assistant feature


Google Assistant can be better at recognizing the voice of users. According to the report by 9To5Google, Lines of code in the new version of Google’s Android application suggest that Google Assistant will get a feature called “Personalized speech recognition” soon. The publication shares a description of the feature, suggesting that it store audio recordings on the device to help Google improve its voice assistant.

In short, Google can store audio recordings of device users and use them to improve hotword recognition technology. As the publication notes, Google “learned how to adjust the mode from the voice data, and sent a summary of the model changes to Google’s servers” which in turn would help improve language recognition across the model and benefit all Google Assistant users.

Google Assistant can be better at recognizing voice commands

The feature enables Google to better understand phrases and words spoken by users after using the hotword “Hey Google.” This can be useful in situations when users create a text message using their voice and other similar tasks. It’s worth mentioning that Google is already using a chip in devices like the Nest Hub and Nest Mini that process users ’most common commands to facilitate faster response times.

In related news, Google is also working on on-device snore and cough detection. Well-known tech publication 9To5Google was able to decompile the latest version of Google Heath Studies. Within the application, the publication is able to locate code strings that suggest the presence of the ‘Sleep Audio Collection.’ As a feature, the Audio Sleep Collection is available for Google Employees only. In fact, the company is working on a component that will collect audio through the microphone of a smartphone for monitoring users ’snoring and cough patterns.

The publication also shares a picture showing the quality of sleep, the time at which the user snores and the time at which the user coughs. “Cough and Snore algorithms” will be used as a bedside monitor for providing users with information related to their sleep. It is known that coughing or snoring during sleep is a sign of bodily diseases, which need to be treated in time.





Source link

Related post