Saw the following post by @tante where they talk about how voice commands have been embedded in our consciousness for years as a really advanced user interface but practically speaking they would be a terrible interface.
“Voice commands” have been embedded into our vision of computing for a long time: Star Trek did it and it does the whole anthropomorphization thing that tech loves.
But from a practical standpoint it’s a bad interface: Think of a handful of people sitting in an office yelling at their computers, it would be like working next to people having loud phone calls all day.
And who exactly wants every thing they do with their computer broadcasted around them?
This is something that has bothered me pretty much since I first watched Star Trek (and how they interfaced with computers) and started working on computers myself. So it was good to see someone else share my point of view on this topic.
From usability perspective, a voice interface is one of the worst ways to interact with the computer, as it is imprecise. The caveat being that it is a good interface from an accessibility perspective but from a general purpose, everyday use I don’t think I would use voice to interact with a computer as it is a very limiting and insecure way to interact with a computer in my opinion.
The way Star Trek thinks about it is that the computer is advanced enough to understand what you are trying to do based on a single line command and then do it for you but we are nowhere near close to achieving that right now and even if the tech existed, imagine sitting in an open office setup where already it is annoying to work because you get to listen into random zoom calls & discussions happening around you. Now think about how bad it would be if we also had to listen to every command that people around you are issuing to their computer (followed by the numerous corrections), I would go mad if I had to listen to it for any period of time.
Another issue to think about when multiple people are working in a small area and trying to interact with their computer: How do you ensure that you are commands don’t accidentally run on your neighbor’s computer? Imagine sitting next to someone logged into a production database while you are connected to a dev database and you tell the computer to wipe all the data in the database. But in addition to the command running on your machine the machine next to you also heard the command and wipes out the entire production database.
In addition there is the whole security aspect of the technology to think about. Do you really want everyone around you to hear what you are working on? (Especially if it is something confidential). Plus authenticating verbally like how they do it on Star Trek is really insecure. In Star Trek when they are issuing a command that requires authentication the person actually just states their authentication code loudly for everyone around them to hear. That is absolutely insecure and easy to bypass. We already have technology that can imitate anyone’s voice really well, so this authentication code could easily be bypassed using similar tech. For example, they could create a Hologram version of the person and have that issue the command or use any of the other similar technology to achieve the same thing. In the non-fiction world, voice modifiers are already good enough to fool most if not all voice verification systems so having a verbal authentication code is not really a very secure way of doing things.
So, while it is cool and looks really futuristic I wouldn’t recommend a voice interface as the only way to work with a computer.
– Suramya