Out of all the voice assistants currently on the market, Amazon’s Alexa seems to have been the breakout star. During the 2017 Consumer Electronics Show, many third-party manufacturers announced that their device will integrate Alexa. Now, Amazon has announced that it is working on more advanced voice recognition technology for Alexa, in order keep its momentum going.
Alexa, the voice assistant integrated into Amazon’s smart speakers like the Echo or Dot, is already quite innovative and includes many features and functions. However, like most technologies, it is far from perfect as many users point out that currently, Alexa has trouble recognizing and responding to requests made by different voices.
In order to address this issues, Amazon is reportedly working on a more advanced voice recognition software which will allow Alexa to tell different voices apart more easily, than what she is capable of currently.
Much like other voice assistants such as Apple’s Siri, Google’s Assistant, and Microsoft’s Cortana, Alexa is meant to respond to one query issued by one person at a time. However, since members of a family can have their own playlists for example when multiple persons ask Alexa to “play my playlist”, it can get confusing.
However, the report reveals that Amazon is currently working on a feature for Alexa which would allow her to match one person’s voice to a specific and unique print for verifying voice commands. This means that Alexa would be able to tell different persons and their voices apart. Furthermore, it can be restricted to only fulfill the requests of only one person and not those of other people.
This feature will prevent accidental purchases made by children who are playing around Alexa, or even accidental requests made when Alexa considers TV audio as the request of its owner. Alexa already has multiple verification measures such as account and PIN input for when it makes a purchase. However, adding a voice-specific layer of security will further improve the service overall.
This type of feature for Alexa is known as Voice ID, and the report reveals that it has been in internal development since 2015.
What do you think about the new Alexa voice recognition technology?
Image source: Wikipedia