Edgar Cervantes / Android Authority
We’ve got some guides on using Amazon Alexa on Android Authority, but you might be curious about the voice assistant’s underlying technology. Here’s a brief explanation of how Alexa works, from its overall structure to how it listens to and responds to voice commands.
How Alexa works: An overview
The basic components of Alexa from the user’s perspective are an Amazon account and an Alexa-enabled device connected to the internet, typically a smart speaker or display. The account allows you to create a profile, save software and hardware settings, and link compatible devices, services, and accessories. Alexa devices listen for voice commands, upload them to Amazon servers for translation, and then deliver results in the form of audio or video. Some models also serve as Thread or Zigbee hubs for compatible smart home products.
All voice commands begin with a wake word that tells a device to listen. The default value is of course “Alexa”, but with the Assistant app for Android or iPhone/iPad you can change this to “Amazon”, “Computer”, “Echo” or (in some regions) “Ziggy”. In fact, the app is practically a third basic component, as it’s required for setting up the device and linking things to your Amazon account.
There are many, many possible Alexa commands, so we won’t dive too far here, but these are natural-language voice queries that cover everything from general knowledge questions to media playback to smart home controls. For example:
Some features require “Skills” to be activated, whether through the Amazon website or the Alexa app. Using the above commands as examples, the middle one wouldn’t work without a skill linking your Spotify account, and thermostat control would require an appropriate branded skill like Ecobee or Nest.
The Alexa app also allows for routines, which is just another word for automations. You can learn more about this in our routines guide. The short version is that they are user created and trigger actions based on voice commands or various conditions like location, accessory status or time of day. For example, a good morning routine could turn on your lights, play NPR news, and warm up your coffee maker via a smart plug when you say “Alexa, start my day.”
To be controlled by Alexa, smart home accessories must specifically support the platform, at least until the universal Matter standard goes live in fall 2022. However, just about every type of accessory is available. Aside from plugs, thermostats, and smart lightbulbs, you can get everything from air purifiers to robot vacuums. These are paired via the Alexa app, regardless of whether they connect via Skills, Thread or Zigbee.
More: How to use Amazon Alexa
How does Alexa hear?
Dhruv Bhutani / Android Authority
While all Alexa-equipped devices have at least one microphone, smart speakers and displays often have two or more. This makes it easier to isolate voices from ambient noise as it produces directional data that can be compared and filtered by signal processing algorithms. There are finite limits, of course – you can’t stand next to a loud TV or dishwasher and expect an Echo speaker to hear you.
Contrary to what you might have been told, Alexa doesn’t constantly record everything you say. It is Constantly listens to its wake word, and subsequent audio (which ends after you stop speaking) is usually sent to Amazon for interpretation. We say normal because Amazon is pushing more and more towards offline processing. However, you’ll need newer devices like the 4th Gen Echo or Echo Show 10, which feature the company’s AZ Neural Edge processor. The feature must also be manually enabled, and devices will continue to upload transcripts.
Amazon says it encrypts audio recordings it uploads, but saves them by default and analyzes “an extremely small sample” of anonymized clips to improve Alexa’s performance. Records have been used in criminal cases, and some sounds or phrases can be misinterpreted as wake words. So if you’re concerned about privacy, you should opt out of saving or regularly clear your voice history. Read our smart home privacy guide for more details and comparisons.
See also: How to set up Alexa for emergencies
How does Alexa react?
The reason Alexa was completely dependent on the cloud until recently is because of the demands of natural language processing. Each command is broken down into individual speech units called phonemes, and these units are then compared to a database to find the closest word matches. In addition, the software must identify sentence structures and terms relevant to different subsystems. When you say “set the thermostat to cool,” Alexa knows to pass this to a smart home application programming interface (API).
This is the main reason why Alexa can distinguish between accents and dialects. There are unique databases for every language Amazon supports, including regional variations, and users must select them in the Alexa app if their device doesn’t come with them pre-installed. An American Echo speaker will not understand German, as anyone asked about Nachtmahr songs will attest.
Machine learning plays an equally important role, as context and history give Alexa a better chance of guessing your intentions. This is why Amazon invests so much in analyzing real-world customer records. Humans tend to use context and history to gauge the meaning of conversations, and with just computer logic, Alexa could interpret something like “Play music by Chvrches” (the Scottish synthpop band) as a request to listen to church choir music. Alexa can and does make mistakes, but the seas of data Amazon has at its disposal means the assistant will evolve over time.
frequently asked Questions
Effectively. While some devices allow offline voice control of volume and hub-linked smart home accessories, or checking and canceling things like timers and reminders, almost everything else requires communicating with Amazon servers and/or linked third-party services. Even devices that can process audio locally still upload transcripts.
Yes, provided you have not muted a device’s microphone(s). It must, in order to respond to its wake word.
Crucially, however, not everything is recorded. Recording only triggers after a wake word is detected, and ends as soon as you stop speaking (or Alexa thinks you did anyway). If you’re concerned about privacy, you’ll need to disable the storage of these recordings or periodically clear voice history.
After some definitions. It’s able to learn and solve problems, such as interpreting voice commands, that it wasn’t preprogrammed to do.
However, it does not display the same flexibility or adaptability as a human or animal mind. You can’t have a real conversation, and learning is gradual rather than on the fly. It’s certainly far from being sentient, no matter how difficult that may be to define.
https://www.androidauthority.com/how-does-alexa-work-3209316/ How does Alexa work? The technology behind Amazon’s virtual assistant, explained