A THIRD GENERATION
IS NEEDED

I long for the good old days when you could call a business and talk to a real person without being screened with endless menu options. I don't like to call a company and hear a recorded voice give me my options that might not even deal with my question. However, the reality is that these answering messages are here to stay.

You can imagine my mild irritation when I called to hear the second generation of these automated answering systems. This new system is equipped with voice recognition capabilities. The automated voice cheerfully says, "If you are wishing technical service, say 'yes!'" It wasn't long before I found myself talking to a computer that seemed to understand me. The machine seduced me into doing something that in another setting would result in a committal for 72-hours in a psychiatric ward.

However, if voice recognition answering becomes the industry standard, allow me to suggest an upgrade to the third generation: emotion recognition. This is both a necessary and a logical next step, because customer service is critical. The corporations that use these first or second generation answering machines must know that they are at best irritating and very depersonalizing. Most of the time when I call one of these companies, I am already upset about their product or I wouldn't be calling. Therefore, it would be prudent for these corporations to do what they can not to incite further bad feelings with their customer base. Bill Gates ought to get on this idea right now.

The software needs to go beyond voice recognition to emotional recognition. For example, I would suggest this third generation of software would be able to recognize not only voice but at least four levels of emotion:

1. Pleasant and happy

2. Annoyed but engaged

3. Pre-ballistic

4. In orbit

Later on, these four could be expanded to even more definitive categories. For the meantime, these four will due for starters. If a customer calls in and the answering machine detects a pleasant emotional demeanor, nothing needs to be done other than merely getting the caller to the person they need.

However, if the person is annoyed but engaged, the program needs to fend off getting the person to the pre-ballistic stage. In this situation, it is vital that what is said and what is played in the background is selected carefully. I'd suggest Rachmaninoff's Rhapsody, Op. 43 as background music. The message should be something like, "I feel that this experience is annoying but let me personally assure you that we are engaged, and soon all your cares and concerns will be satisfied successfully. Please bear with us as we determine the office that you need to reach."

If the emotions recognition detects a person about to go off, the verbal response and background music needs to be different. For example, "Boy, let me tell you, I had the same problem that you are experiencing. It was all that I could do to avoid going off on the person in tech support. However, I feel your pain and want to assure you that in no time, you will be up and running. Just bear with us a bit longer and all will be fine." The music should be something soothing, like Tchaikovsky's Mozartiana, the 3rd movement.

However, if the software picks up that someone has already entered low earth orbit, that vocabulary and music wouldn't be appropriate. The ballistic person needs far more sensitivity. "While you are waiting to talk to a real person, talk to me about your feelings, alienation, and anger. I'm here to help, just get it out. It won't work unless you talk about those feelings. I'm here to help and listening is a part of what I do. Go ahead, get it off your chest." Then every fifteen seconds have the software say, "Hmm. I can see why you are so upset." The background music should be very mellow-something like Robert Schumann's Traumerai.

Automated answering systems are the wave of the future, embrace them. However, it will be a lot easier if corporations would go to a third generation of these devices as soon as possible. The emotion recognition devices should hold us until the fourth and final generation is invented-mind-reading. Mind-reading would eliminate all those canned voices altogether.