Amazon’s Virtual Assistant
Responsive robots may have begun as a dream and subsequently a Jetsons character but as we get closer to the broadly defined sci-fi “future”, more and more of those futuristic ideas are becoming a reality. Alongside Apple’s Siri, and Microsoft’s Cortana, the Amazon Echo features a similar computational companion, Alexa.
The Echo includes a Texas Instruments ARM Cortex-A8 processor, 256MB of LPDDR1 RAM, and 4GB of storage space while also utilizing cloud based processing and storage through Amazon Web Services. In its default mode, the device is constantly waiting for its name or customizable “wake word” but can also be configured for manual activation. Once it receives the go-ahead, music, audiobooks, news, calendars, and even the functionality of a home such as lights and thermostats are at the user’s command. With a 7 piece microphone array and omnidirectional audio, its clear the product was designed to hear, and be heard, from anywhere in the room and its voice recognition has had much success in command interpretation despite interference such as running water, or a loud TV.
Limitations with some of these features however, do exist. As expected with many interpreters of the English language, word confusion does still occur and Echo is sometimes mixed up when presented with certain homophones. Additionally, certain background noise may not be viewed as such as seen in an ironic set of incidents for NPR listeners when a segment discussing how the voice activated assistant was helping customers. Upon hearing its wake up word, the story ended up triggering Amazon Echos in the homes of a few listeners and modified or reset their thermostats.
Additional concerns revolve around privacy and security to which Amazon responded to swifty. Caution around the device’s feature to perpetually monitor for its wake up word was dissuaded by the online retailer when they enumerated the absence of any streamed recordings prior to the wake up word. When, and only when, Echo receives this cue, does it begin to stream audio to the cloud for processing. As for Echo’s future, updates and apps are not in short supply. Because much of Echo’s intelligence lies in the cloud, significant functional enhancements can be made to Echo without updating the software version it is running. New features and fixes will be implemented without the user needing to update the product so as time goes on, the bridge between reality and the sci-fi fantasies of the past looks to be growing shorter and shorter.