Voice UX

Voice UX: Pixi-stories on command

Pixi-story

A bedtime story is part of the bedtime ritual for many families – and Alexa can also be helpful here and come up with a bedtime story on command. But in order for the experience to be approximately as beautiful as it is when read by parents, there are a few things to consider. The most important consideration in advance: the target group of a bedtime story skill is primarily children and this is exactly what you have to do justice to in UX design.

Pixi-story

alexander dummer_unsplash

Here are the most important design rules:

First. Prompt design.

The cognitive load that is widely discussed in the voice sector and that can be expected by listeners without overwhelming them. And it is even lower in children than in adults. A great deal of attention should therefore be paid to the prompt design.

One technique that is often mentioned in this context is the so-called “one-breath” technique: if a prompt can be articulated in a single breath, it is short and concise enough so that children are not overwhelmed. In addition: if the child is to react to a prompt with a voice input, it is important to position a specific question at the end of a prompt (rhetorical questions are not recommended, even if they bring an element of ease into the dialogue – children will definitely try to answer and may be disappointed with the failure of their attempt).

Secondly. Sound design.

A good way to create lively dialogues is to use child-friendly sound effects and voice recordings by professional speakers. As good as Alexa’s voice has now become, in terms of intonation and emotion, it is still not comparable to human speakers and therefore less attractive and captivating for children. Voice recordings, however, bring a negative factor into the design process that should not be underestimated, as they – once recorded – allow little or no flexibility for changes in the design flow. A playful dialogue between Alexa and a recorded speaker can be the solution here. This way, you can react to the unforeseen by simply handing it over to Alexa in a dialogue.

An example: in a bedtime story skill, new content that has not yet been produced at time X should be added periodically. A speaker can therefore not record all the story titles or would have to do this over many sessions. Here a playful dialogue between the speaker and Alexa can take place:

Third. Choices.

For Alexa as well as for children, it makes sense to choose between different options in the skill, to choose these options in such a way that they differ from each other as much as possible. This prevents confusion and improves the UX. For example, if you give children the option to choose between two bedtime stories, it makes sense to choose these stories according to the criterion of auditory distinction and to avoid words repeating themselves.

The examples show another principle that must be observed when dealing with content: if several options are offered, they should belong to different categories in order to do justice to the diversity of the target audience – not every child wants to hear a story about vehicles as other children might find animal stories boring.

Fourth. Fail with Grace.

Although this should of course also be taken into account in adults, it is even more important for children not to let a skill run into a dead end. Following the “Fail with Grace” principle, there must always be a result at the end of an interaction. Where adults understand that technology may not yet be able to keep up with their own expectations, children quickly lose interest and do not open the skill again. Good error handling is crucial here and should open up many possible paths, all of which lead to a satisfactory result.

An example:

Designing for children is a very unique design experience and requires a specific and complex approach to the topic of UX in order to do justice to this special target group. But one thing can be promised: once everything works, you will be as happy as a child.

Share via 

Bias in AI in 2020

Multilingual Smart speakers in 2020

  • Julia Pflueger
  • AI NLP

Voice UX: Pixi-stories on command

All entries