Amazon upgraded its Fire TV voice search feature, recently adding phonetically blended results (PBR). Since AI voice recognition isn’t perfect and doesn’t account for accents as well as unintentionally pick up other sounds in the room, PBR helps eliminate repeated attempts to voice search for specific programs by offering up a list of related search results.
“Previously, Alexa was not tuned into individual Fire TV customers’ preferences. That’s the layer my team is adding. We are connecting Alexa machine learning with Fire TV search algorithms to build toward an end-to-end algorithm to help customers find what they’re looking for,” said Kanna Shimizu, Senior Manager of Research Science for Alexa AI’s Natural Understanding Group. “Let’s keep the door open to different interpretations of what the customer may have said, so they can decide for themselves on the search results screen.”
Although Amazon Fire TV’s PBR voice search probably will bring up a host of titles unrelated to your search, it is attempting to provide the series or movie you’re actually searching for among the offerings.
As stated in an announcement yesterday by Amazon, “It works by harnessing a suite of Alexa machine learning (ML) models to generate additional, similar-sounding words to inject into Fire TV’s search function to broaden the scope of the results presented to the customer… About 80 percent of the 20 million or so unique search terms that Fire TV deals with are augmented by PBR.”
PBR works with Amazon Fire TV’s ReRanker search algorithm, so searching for Encounter may bring up Enchanto as well. This is a slight improvement from previous versions of Amazon Fire TV voice controls, which would bring up one or the other and require users to repeat the process or resort to typing in their search.
Alexa and PBR algorithms work together to learn the viewing habits of subscribers to (hopefully) improve Amazon Fire TV’s voice control features and get search results right the first time.