Potentials of Voice Interaction in Industrial Environments

Hi. Both “Speech/VUI” and “Non-speech” have pros and cons. But many still found that even though VUI is very straight forward and easy to understand, “Non-speech” is better since it can be delivered much faster. Agreed with the phrase cited in your article “speech is the bicycle of user-interface design: it is great fun to use and has an important role but it can carry only a light load”. Great findings. Cheers.
Faieza Abdul Aziz

In your introduction, you mention phone ticket booking systems as becoming more widespread. Not more popular though. I think that received opinion is that such things are awful. This may change as the technology matures and the system makes fewer mistakes.
However, I agree that there VUIs will only be useful for some applications. There's a question of feedback. You can tell when you've turned some knobs, a small arrow is the only display required. Not so clear with voice. If a command turns the lights up and down, that's clear. If it turns a kiln temperature up and down, not so clear. Currently, the VUI tends to repeat the command back to you. For some, that's just frustrating.
What you say about earplugs is telling I think. If VUIs can be incorporated in situations where the user doesn't notice the difference, then they will succeed. A user interface should be for the convenience of the user. Voice booking systems are implemented for the convenience of the company, so that they don't have to employ more staff...










You present some very interesting advantages, mainly the hands free advantage, and the ability to alert the user focus. There is one problem that is not mentioned though.This is that the industrial environments are also involving risk in many situations; In the case of voise interaction the user is not the one who stands in front of the machine/computer but anybody in the proximity defined by the range of the acoustic sensor. In this case the danger of wrong data/commands being issued by non qualified/authorised people increases. This would require some type of voice authentication, is something like that possible ?