Voice is (mostly) Bad UI

🌈 Abstract

The article discusses the limitations of voice as a universal user interface, arguing that it is not poised to take over as the primary mode of interaction with technology.

🙋 Q&A

[01] Voice as an Interface

1. What are the key reasons the author believes voice is not the future of user interfaces?

• Voice is incapable of representing abstract thoughts and the subconscious mind, which are crucial for activities like writing and research.
• Voice can make many tasks harder, not easier, as it requires constant back-and-forth to resolve ambiguities and preferences that are easily handled through a visual interface.
• Voice is privacy-invading, as people may feel self-conscious about making requests out loud in public.
• Visual interfaces help users understand the limits of software capabilities, which may not be as easily conveyed through voice alone.
• Users don't want to be talking all the time to accomplish everyday tasks, as many of these are better suited to familiar interfaces.
• Voice interfaces cannot easily resolve real-world ambiguities that can be more readily handled through visual feedback.

2. What examples does the author provide to illustrate the limitations of voice interfaces?

• Ordering an Uber or food delivery is more complicated through voice, as it requires resolving details like vehicle type, pricing, and menu options that are more easily handled through a visual interface.
• Checking messages or performing other routine tasks may feel awkward or intrusive when done through voice in public.
• Voice interfaces cannot easily convey the limitations of software capabilities, such as the inability of Google Maps to provide routes based on scenic criteria.

[02] Limitations of Voice Interfaces

1. What are the key reasons the author believes voice is not well-suited as a universal interface?

• Voice is limited in its ability to represent abstract thoughts and the subconscious mind, which are crucial for activities like writing and research.
• Voice can make many tasks more complicated, not easier, as it requires constant back-and-forth to resolve ambiguities and preferences that are easily handled through a visual interface.
• Voice is privacy-invading, as people may feel self-conscious about making requests out loud in public.
• Visual interfaces help users understand the limits of software capabilities, which may not be as easily conveyed through voice alone.
• Users don't want to be talking all the time to accomplish everyday tasks, as many of these are better suited to familiar interfaces.
• Voice interfaces cannot easily resolve real-world ambiguities that can be more readily handled through visual feedback.

2. What examples does the author provide to illustrate the limitations of voice interfaces?

• Ordering an Uber or food delivery is more complicated through voice, as it requires resolving details like vehicle type, pricing, and menu options that are more easily handled through a visual interface.
• Checking messages or performing other routine tasks may feel awkward or intrusive when done through voice in public.
• Voice interfaces cannot easily convey the limitations of software capabilities, such as the inability of Google Maps to provide routes based on scenic criteria.