Customizing Voice Experiences for Echo Auto and Other In-Vehicle Devices

After a year-long, invite-only release of Amazon’s Echo Auto, the device is available for all to purchase. Amazon received over one million pre-orders for Echo Auto and Nationwide Insurance has announced it will distribute the device to another million drivers that are its customers. In addition, Voicebot data show there are as many as 50% more monthly active users of voice assistants while driving as through smart speakers. Now is the time for Alexa developers to start creating in-vehicle voice experiences. However, there are some new values you should be aware of that can be used to customize Alexa Skills for the car.

Alexa Devices for the Automobile Differ in Capabilities

While the Echo Auto is Amazon’s flagship product specifically made for automobiles, some users are bringing Alexa to the car using the Echo Input or Dot, their mobile phone with the Alexa App, or one of the third-party products from makers such as Garmin and Anker.

For this discussion, we are not considering the hardware aspects of these devices that make them special-built for automobiles. These aspects would include such things as the number of mics or whether the device connects to your car’s audio system via AUX, Bluetooth, USB, or FM Transmitter.

The following table shows what values are available on various devices for developers of custom Alexa Skills as they create in-vehicle experiences. Echo Auto today, offers two values that can be particularly useful for automotive use cases but are not available through third-party devices.

Custom Values for Alexa Auto Skills

Let’s go through each of the values that Alexa Skill developers have that allows them to customize the experience for a single skill that all family members will use in whatever car they are driving.

User ID – the user ID is a unique identifier for the registered Amazon account that is active on the device and initiates the request to the custom skill. Each skill enabled for an account will have a different user ID, but all devices using that skill for a single user will have the same user ID. Store information with this user ID as the key if you want it available for any user on any device. This can also be used for those family members who haven’t set up a voice profile.

Voice Profile (Personalization Preview) –Voice Profiles have been around since October 2017 and allow each member of a household to train Alexa to recognize their voice. After answering a series of prompts, you can ask “Alexa, who am I?” and Alexa will identify your profile. The Skill Personalization feature is currently in preview, but it will allow custom skills to identify those users that have a voice profile. Presumably, those without a recognized voice profile could still have values stored at the account level with the user ID. Imagine a in-vehicle skill recognizing the driver and their preferences regardless of which car they were driving. Maybe developers will one day have access to the user’s profile name so that Alexa can talk to the driver by name like KITT talking to Michael on Knight Rider.

Device ID – The device ID uniquely identifies a device. In a home scenario, you could configure a skill to recognize that the device was in the kitchen, bedroom, family room, etc. For those vehicles that have a dedicated device you could configure the custom skill for the car’s make, model, year, or VIN. Or, when the skill is first invoked, you can assign values that indicate ownership or nicknames such as: “Mom’s car”, “Silver bullet”, or “Old guzzler.”

Supported Interfaces – Each device passes a list of supported interfaces that can inform a custom skill of features supported for that device. Two of those values are AudioPlayer and Geolocation. These are included as part of the request and indicate if the device supports playing audio files and has access to latitude and longitude respectively.

Audio Player – A key usage of Alexa-enabled devices is to listen to music and podcasts. This will likely also be true for the car. With this flag, a skill will know if the device can play audio content.

Location Services (Geolocation) – Besides navigation, location services can be used to get the real-time location of the Alexa-enabled device. This is different than the zip code or address you can assign to a stationary device in your home or office.

Automotive property – Currently, the only Alexa-enabled device that passes the Automotive property is the Echo Auto. Amazon has a design guide specifically for in-vehicle voice experiences. For example, it is recommended that any interactions while in the car should be shorter and be free of distracting or repetitive noises.

Multi-Car, Multi-Device Scenario

One unique aspect of automotive use cases is the potential value of geolocation. Echo Auto introduces this service for the first time along with the ability to uniquely identify an automobile. Here is a quick scenario to illustrate how these values can be used.

Imagine a family with two adults and two teen drivers and three vehicles. The adults each drive their own cars and the teens share the third car. But there are times when the adults or the teens drive any of the cars. Two of the cars have an Echo Auto device and the teens use their mobile phones and the Alexa App in the third car.

The skill we are designing is called Ultimate Car Spot. It will tell you where you parked your car. Family members Mom, Dad, and Rachel have set up a voice profile. Ben has not. When the skill is invoked for the first time on each of the devices, each user set up a nickname for the device. The Echo Auto in Mom’s car is “Silver bullet.” Dad’s Echo Auto is “Thor.” Rachel’s mobile phone is “Pikachu” and Ben’s phone is “Ben Phone”.

Mom takes her car to a concert and once she is parked says, “Alexa, tell Ultimate Car Spot that I am parked.” Using the Automotive property, voice profile, device ID, and location services the spot is remembered, and Alexa responds: “Mom parked the car, Silver bullet, at <geo location>.” If Rachel parks the teen-shared car and invokes the skill then the lack of the Automotive property and the existence of the voice profile, device ID, and location services, Alexa would respond: “Remembering the location of Pikachu for Rachel at <geo location>.” If Ben were to issue the same command with his phone and since he doesn’t have a voice profile, the response would be: “Remembering the location of Ben Phone at <geo location>.” You can imagine the various permutations of responses depending on which driver used which car and device.

If all these settings for the skill were persistent using the account’s user ID then you could ask questions to the skill such as “Where is the silver bullet?” or “Who last parked Thor?”

On-the-Go Alexa Skills to Rise

As you might expect, the Echo Auto is the only device at this time that includes all the current features that developers can use to customize an in-vehicle voice experience. Third-party devices such as the ROAV Viva and Garmin Speak as well as Amazon devices Echo Input and Dot are not able to execute the example scenario listed above today since it requires Location Services and for part of the use case, the Automotive Property. I suspect these features will soon become available on other third-party devices that are specialized for the car, but for now they are limited to Echo Auto.

Regardless, developers and designers can start creating new skills for users interacting with Alexa devices in their cars. Voice assistants like Alexa are no longer just for the home and on-the-go scenarios have the potential to significantly expand the value for users. Echo Auto is the latest catalyst to help developers provide unique, customized experiences for on-the-go use cases.

About the Author: Mark Tucker is a Senior Architect, Voice Technology at Soar.com and an Alexa Champion with over 20 years of software design and development experience. He is the organizer of the Phoenix Alexa Meetup and the Phoenix Chapter of the Ubiquitous Voice Society. You can find him on Twitter @marktucker or LinkedIn.