Summary
Designers must deal with the current state of technology, which presents issues of misrecognition and misunderstanding in speech recognition. They must also deal with the “robotic” tone to synthesized natural language voices that needs refinement in syntax and prosody. For both input and output, technologists have yet to get the pace of speech to truly match natural human tempo. These issues can be mitigated through usability testing, iterative design, and thoughtful use of natural language.
As the state-of-the-art in natural language improves, as well as the quality of speech recognition and synthesis, these technologies will become more prevalent and effective. However, even today speech technology is commercially available and useful for both niche and general markets. The current state of computer software is a litmus test for the status of speech technology and is discussed in the next chapter.
Get Designing Effective Speech Interfaces now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
