Alexa Conversations to Automate Elements of Skill Building Using AI and Make User Experiences More Natural While Boosting Discovery
Amazon announced Alexa Conversations at the re:MARS conference yesterday in Las Vegas which is a new tool for Alexa skill developers designed to streamline development and make skills more easily discovered by users. The new capability is being presented at the conference and in Amazon’s Alexa blog as a tool for building Alexa skills with “less effort, fewer lines of code, and less training data than before.” Rohit Prasad, vice president, and head scientist, Alexa artificial intelligence, said in his keynote presentation that Alexa Conversations enables the development of Alexa skills with three times fewer lines of code and ten times less data. Prasad expanded his public comments in an interview with Voicebot:
Within the Alexa Skills Kit, [Alexa Conversations] is one module, or one tool chain, that lets you automatically induct the dialog flow instead of hard coding it…there is a new capability to estimate the dialog flow automatically rather than a developer writing all possible structures.
Improving Alexa Skill Discovery
In addition, at a breakout session reviewing the new capability, Sanju Pancholi, senior product manager, Alexa AI, also indicated that skills using Alexa Conversations will also be eligible for new types of organic discovery. When a user makes general requests about a task, skills built using Alexa Conversations may be automatically invoked for fulfillment, without the user needing to know the skill name or even that it exists. Ashish Shah, CTO of Pulse Labs commented:
“New features coming from Alexa Conversations represent an exciting new development for both developers and end users. This could help improve discoverability and usability while decreasing development work for third-party skill developers.”
Dramatic Code Base Reduction
Eric Posen, principal product manager at Atom Tickets, was part of a demonstration preview along with OpenTable and Uber. Posen said in a presentation that Atom Tickets had attempted to build an Alexa skill last year, but it turned out to be too complex saying, “when we launched it, we weren’t impressed.”
However, using Alexa Conversations was an entirely different experience because the AI was able to automatically define much of the dialog flow required to purchase a movie ticket along with many of the common utterances. His team was not required to define those elements using Alexa Conversations. A flow diagram of the earlier Alexa skill had up to 13 steps, while in the Alexa Conversations version he depicted just four steps in the Atom Tickets skill to complete the purchase. In this instance, Alexa gathers information from the user about their preferences in advance of passing the experience along to Atom Tickets for fulfillment.
A Different Development Approach
Steven Arkonovich, an independent developer with the most popular third-party Alexa skill for weather queries, Big Sky, echoed a common sentiment among many developers that the change will be significant. To Archonvich, it appears Alexa Conversations could eventually shift skill development from an imperative approach where all details of the user experience are defined by the developer to a declarative model where the logic is defined but not all of the control flow. Arkonovich told Voicebot in an interview:
“My initial impression is that Alexa Conversations is very promising and, depending on the details of implementation, could alleviate several major problems such as context, discovery, and invocation names. Some aspects of ‘conversation’ won’t require anything from the developer, likely. For example, which other skills your skill might be connected to—Alexa will decide what overall plan you [as a user] might be trying to implement with when you make, say, a weather request. But in terms of building skills, my sense is that maybe not right away but eventually, it will be a very different skill building experience, away from imperative programming toward declarative programming. They kept talking about a ‘new paradigm,’ so I think the changes are going to be significant.”
John Gillilan, founder of Bondad.fm, also sees Alexa Conversations as ushering in an important change for Alexa skill developers with unclear implications. He told Voicebot, “Without having seen any details beyond the presentations, my first impression is that today marks the start of a new era of what it means to build for Alexa. It feels like the current paradigm of interaction model design will soon feel very quaint.”
Who Controls the User Experience?
Some developers did express concern that handing much of the dialog flow and sample utterance creation over to Alexa’s AI may save time but also could result in less control over the user experience. There will be aspects of skill experience that developers may not be able to predict. In addition, it raises questions about how to develop test scripts if some of the functional performance of a skill is not known by the developer. Prasad responded that these concerns can be addressed with tools that will be provided to developers saying:
“Developers can always add more control…Existing things like adding logic that overwrite the models will be brought in. That already exists. This is why we are going into preview so we can learn what level of flexibility developers want at their end or what kind of controls they want…Think about very system initiated dialogs like games. You want way more control as a developer than the user; like if you are playing Jeopardy! or you are playing 20 Questions. So, in that case, the mix of this automated flow induction versus the hand supported or structured more from the developer side is different…Instead of writing code, you are just simply creating a few sample data of interactions with your API, which means the onus of code writing has gone down. You still control a lot of the experience the same way…The burden has shifted from code authoring to data creation…The cognitive burden on the developer is going down.”
Who Will Use Alexa Conversations and When?
It is noteworthy that the feature benefits behind Alexa Conversations will not apply equally to all skill types. Prasad noted, “In certain scenarios like very structured games where you want complete determinism…where you want certain actions to always happen in a certain order, you may not use this…I think anything which has a lot of different steps in completing an action—there are many of these skills like Expedia, they built a massive skill with so many different examples and many different hand codings. Those are the right ones to immediately get on and the other end of the spectrum is very directed games and that I think may want to not have ambiguity of any sort.”
Whether you consider this a significant shift in Alexa skill development or simply a way to reduce the lines of code required to achieve a result, Alexa Conversations as a development model will not be common anytime soon. Developers that are interested in using Alexa Conversations may apply to use the tools during a preview period. It will likely be some time before Amazon has enough developer feedback, production data, and new tools to support the model for Alexa Conversations to become generally available. If successful, however, it will likely usher in the most significant change in Alexa skill development since the introduction of Alexa Skills Kit in 2015.
Follow @bretkinsella Follow @voicebotai