The COO of Parlance Corp., Joseph Maxwell, recently outlined the five major components to a successful speech product for an auto attendant. It’s a good summary, so let’s take a look at it.
Successful speech-enabled call routing tools, he says, are made of components working together to connect the caller to his destination “according to a spoken name or phrase.” That being the case, here are the five major components you want:
The speech recognizer. This is what acts as the interpreter for matching a spoken name or phrase with an entry in the system’s directory. “Often disparaged in contemporary society (see any number of comic strip plots pertaining to poor speech recognizer performance for reference),” as Maxwell notes correctly, it’s in a thankless position -- it’s where the user interacts with the application, and as such gets the abuse when the system doesn’t function the way the user expects.
The grammar. This helps determine the spoken request for the purpose of connecting the caller accurately. Formed well, it defines for the recognizer the expected words, pronunciation and grammatical structure of the request.
And because no matter how comprehensive the dictionary you can put in one of them, they’re never going to have 100 percent of the possible pronunciations for all names, so as Maxwell says, “attention must be given to pronunciations for the recognizer to correctly listen and make a match with what the caller requested.”
The directory. The goal here, Maxwell explains, is to “minimize gaps in the system between what a caller might request and what destinations are actually available in the directory.” He notes that identifying and including all possible caller requests -- all the employees, departments, contractors, vendors, what have you -- does increase the chances that the caller will find who he’s looking for, and will use the system again, instead of zipping straight to the operator next time he calls.
The routing table. As you might expect, this is the guts of your operation. It doesn’t matter how well the rest of it is operating; if the routing table hasn’t been set up correctly, the call’s not getting to the right person. It also should direct the system to the least-cost-routing dialing pattern for each destination for cost savings, Maxwell says.
The dialogue: Done correctly, Maxwell says, “it needs to give specific instructions about what the caller should request,” such as first and last name, name of the department, state and whatever. Oh, and importantly, he says, be specific and as terse as possible, “or else the caller will become frustrated and simply zero out to the operator or hang up.”
So it’s not just the speech recognizer that needs to be up to snuff. Even if that’s working perfectly, if the router or dialogue aren’t right, it doesn’t matter how well the system recognizes what’s said to it.
Want to learn more about the latest in communications and technology? Then be sure to attend ITEXPO West 2011, taking place Sept. 13-15, 2011, in Austin, Texas. ITEXPO (News - Alert) offers an educational program to help corporate decision makers select the right IP-based voice, video, fax and unified communications solutions to improve their operations. It's also where service providers learn how to profitably roll out the services their subscribers are clamoring for – and where resellers can learn about new growth opportunities. To register, click here.
David Sims is a contributing editor for TMCnet. To read more of David’s articles, please visit his columnist page. He also blogs for TMCnet here.
Edited by Tammy Wolf