DTV Voice
2017 | UX | UI | STB
The aim of this project was to align with the competitive landscape and offer DIRECTV customers a voice-controlled option. To achieve this, we partnered with Amazon Alexa, leveraging this smart home device as a bridge to manage content on their Set Top Box (STB).
Understanding that voice interactions are natural behaviors, we took a human-centered approach into comprehending how our users engage with voice applications. Engaging in conversations with users helped us outline the various command types and understand common phrases used to control DIRECTV. I synthesized these insights into a comprehensive Voice Framework, linking user inputs to onscreen outcomes.
01 - MY ROLE
Serving as the primary UI/UX Designer on this project, my role encompassed regular meetings with stakeholders to provide updates. This ensured everyone was aligned with the design team's vision and project goals. Given the ambitious timeline, which required constructing the voice framework from the ground up, the process was notably demanding. Additionally, I consistently attended demos to promptly address any design or development concerns.
01 - MY PROCESS
research —> competitive analysis —> wire-framing
—> user testing —> executing final design
Users testing provided valuable feedback that found out that a voice language must be designed in a way that follows the natural voice language that users are familiar with and must be adaptive to the various types of voice utterances people use. Which then led us to making the following hypothesis…
“how might we design a voice system that accommodates the natural voice behaviors of our users in a seamless and intuitive manner?”
During this process I met with the research team to understand & synthesize their findings into our design and conducted user testing. I utilized competitive analysis to inform decisions and understand how our competitors were able to use voice effectively and learn from systems that work already. Ultimately the biggest challenge with voice is that if it is not designed well, users lose trust in the product. I studied Fire TV, Roku, Apple TV, as well as Gaming modules that use voice.
02 - DESIGN REQUIREMENTS
After the research and brainstorming phase was completed, I was able to translate the work into the following design intents & requirements…
Translate Intent & Thought - Implicit and explicit communication need to be understood.
Affirm Input - When we ask one another a question, we must acknowledge that it was understood
Bridge Into Reality - Once translated, communication needs to actually invoke something
Efficient Gratification - We only utilize speech when it truly benefits us
03 - DEVELOPMENT & PRODUCT CHALLENGES
The development schedule for the product was tight so not all features were implemented in time for launch. This resulted in a product that was incomplete. In hindsight, the product needed more time to be built and should not have launched until all of the features were addressed. There were also cross platform disconnection points that were a result of the DIRECTV remote not having its own voice button & utilizing a 3rd Party Voice component.
In order to better understand platform disconnection points, we created User Journeys based on the following command intents. Research helped us synthesize the following 5 user command intents, which we used to construct our Voice Framework.
• Navigation: Content, channel (tune) and UI destination (“Alexa, go to ESPN”)
• Play: Launching full screen content (“Alexa, play Game of Thrones on HBO)
• Record: Recording content (“Alexa, record Season 3, Episode 2 of Breaking Bad”)
• Trick Play: Controlling full screen video (“Alexa, Pause”)
• Search: Finding or exploring content (“Alexa, Search for Godfather”)
For each of the command intents, I identified synonyms, mapped out user journey, and created a results matrix of what UI to display for different cases.
05 - EXPECTED SYSTEM BEHAVIORS
Overall, the System will prioritize taking the user to a content’s Info Screen. The system will directly tune if the user specifies “Play” as their intent for a specific piece of content available within the library or uses the “on now” modifier.
• “Alexa, Find Game of Thrones on now” —> Directly Tunes
• “Alexa, Play Game of Thrones season 1, episode 4” —> Directly Tunes
In addition, for each result the system is processing through Alexa, the system will display branded utterance to differentiate between commands triggered by the remote vs. a 3rd party app. Alexa should also have a specific response based on successful, misunderstood, or unaccepted forms of Command, Content and/or Modifiers. We have also provided potential error cases for what is out of scope.
07 - UX DOCUMENTATION
07 - VOICE UI
Voice Overlay - when captures user’s utterance on screen…
Voice Results Screen - If the utterance matches multiple results