Amazon announced that its artificial intelligence assistant Alexa will have the ability to automatically identify the user’s voice. When the user utters a voice command “Alexa, play music”, the system will select personalized music that suits the user’s habits from the Amazon music platform.
Setting Alexa speech recognition profiles can play personalized music for users, push personalized messages, and allow users to purchase without having to enter a four-digit PIN (voiceprint?). Of course, this automated voice recognition feature can be disabled in the Alexa app.
In October last year, for the first time, Amazon had publicly allowed Alexa to have voice recognition capabilities. After that, Amazon had been trying to make it easier for people to create their own unique voice profiles. Initially creating a speech recognition profile needed to operate on the Alexa app. In February of this year, Amazon updated the ability to use Echo speakers to create profiles.
At present, we only know that the process of creating a speech recognition profile requires the user to say 10 specific phrases. An Amazon spokesperson refrained to provide more detailed information on training Alexa to recognize speech.
In addition to providing personalized music, the Amazon team is also conducting further research and development to better understand the user’s emotional state when talking to Alexa, which can bring more personalized experiences in the future.
Also, the developers can now add 8 kinds of unique anthropomorphic sounds for Alexa from AWS Polly, which are very different from Alexa’s existing sounds. Earlier, Amazon has added voice purchase capabilities to Alexa so that third-party developers and businesses can accept payments from millions of Alexa users.