Voice Content and Usability

We’ve been conversing for a long time. Whether to present information, perform transactions, or just to check in on one another, people have yammered aside, chattering and gesticulating, through spoken discussion for many generations. Only recently have conversations started to be written, and only recently have we outsourced them to the system, a system that exhibits a significantly higher affinity for written communications than for the vernacular rigors of spoken language.

Laptops have trouble because between spoken and written speech, talk is more primitive. Machines must wrestle with the complexity of human statement, including the disfluencies and pauses, the gestures and body speech, and the variations in expression choice and spoken dialect, which may impede even the most skillfully crafted human-computer interaction. In the human-to-human situation, spoken language also has the opportunity of face-to-face call, where we can easily interpret visual interpersonal cues.

In contrast, written language develops its own fossil record of dated terms and phrases as we commit to recording and keeping usages long after they are no longer relevant in spoken communication ( for example, the salutation” To whom it may concern” ). Because it tends to be more consistent, smooth, and proper, written word is necessarily far easier for devices to interpret and know.

Spoken speech has no such pleasure. There are verbal cues and outspoken behaviors that mimic conversation in nuanced ways, including how something is said, never what. These are also included in conversational cues that emphasize and enhance emotional context. Whether rapid-fire, low-pitched, or high-decibel, whether satirical, awkward, or groaning, our spoken language conveys much more than the written word had ever muster. So as designers and content strategists, we face exciting challenges when it comes to voice interfaces, the machines we use to conduct spoken conversations.

Voice Interactions

We interact with voice interfaces for a variety of reasons, but in The Conversational Interface, Michael McTear, Zoraida Callejas, and David Griol claim that these motivations largely reflect the reasons we engage in conversations with other people as well ( ). Generally, we start up a conversation because:

• we need something done ( such as a transaction ),
• we want to know some