In March 2021, Google started using the unified learning method on the Android platform to develop the “Hey Google” command. Upon that development, it was learned that Google is working on a personalized speech recognition feature for the assistant.
According to the information received, it now seems that Google Assistant will become better at recognizing frequently used words and names. According to the strings in the latest versions of the Google application on Android, a description given in the Google Assistant settings has been made.
Assistant will recognize your voice better by looking at the voices you recorded earlier
The description in Google Assistant’s settings is exactly as follows: Store the voice recordings on this device to help you get to know them better. Audio remains on this device and can be deleted at any time by turning off personalized speech recognition. Learn more”
This can be linked to an existing support article on Google’s use of unified learning to enable users to enable custom word using voice recordings stored on their device to refine commands like “Learn more”, “Hey Google” detection .
This feature for Assistant aims to extend these machine learning-based improvements beyond the “Hey Google” command to your actual Assistant commands, specifically names and common words. The audio of past conversations will be stored on the device and analyzed for future transcription.
It is not clear when this feature will be available or how much of an improvement it will have. What are you thinking? Please do not forget to share your thoughts with us in the comments section.