Amazon Alexa Gains Voice Identification To Recognize Who's Talking
Amazon has updated its voice assistant Alexa with a cool new feature that many people who have an Echo device with multiple users have been wanting. That new feature is the ability to identify voices and allow access to your own messages, playlists and other content without having to tell the Echo device specifically which content to access. The new feature started to roll out yesterday.
Voice recognition works with the Echo smart speaker and any other devices with Alexa built-in. The feature comes after Google launched its Home speaker earlier this year with similar functionality. Setting the new feature up requires the user to register their voice, and once registered you can sync it with Amazon account preferences.
There is a caveat to the voice recognition feature at launch, however; voice recognition only works with certain features and services. It will work with shopping, Amazon Music, Flash Briefings, and Alexa-to-Alexa calling. Amazon is reportedly planning to roll out voice recognition support to other services and apps in the future.
For instance, to be able to access your own playlists on Amazon Music, you need to train Alexa on your voice and then pick your Amazon Music account on the screen after training. Each of the family members on the account can register a voice and then simply say "play music" and your own lists will play.
For shopping, users will no longer need to speak their voice confirmation code to confirm purchases. Alexa is now able to recognize your voice for Flash Briefings and automatically skips stories you listened to from your feed. You do have to go into the app and set this up.
The training process does have you speak some specific phrases to train Alexa on your voice. Other family members can train Alexa on their voices using the Alexa app and by signing in with the Amazon account used to set up the Alexa device.It should be noted that some people are very concerned about privacy with this sort of voice assistant because what you say is recorded and sent out to servers to be analyzed. Google recently had to pull a feature from the Home Mini assistant because it was constantly listening to what was being said.