Thursday, July 15, 2010

There's more to biometrics than David Caruso

Yes, I know that I just published a post that states that biometrics isn't quite as glamorous as TV makes it out to be.

But perhaps this headline will catch your eye: Zetes Biometric Voice Solution Provides ROI in 6 Months to European Lingerie Sector.

OK, what does voice biometrics have to do with lingerie? And was Barry White involved? No, he wasn't:

Zetes’s voice solution 3iV Crystal is used for automating the picking and cross-docking processes at VFB Lingerie Europe's main warehouse in Igualada, Spain, which has an area of 8,000m2. All the company’s multinational brands are distributed from there. The Igualada facility can send out anything from 15,000 items a day to 25,000 at its busiest times.

3iV Crystal was integrated directly into the warehouse management system already used by VFB Lingerie Europe (Manhattan WMOS), and did not require any middleware. Zetes also adapted its voice solution to meet the company’s needs. Since the company has a wide range of products and a high volume of information, Zetes customised the voice system so that the company could conduct picking and cross docking tasks using voice technology. The system performs various tasks simultaneously, and manages all shipments, taking into account styles, sizes, colours, etc, and obtaining real-time information whenever required.


So has voice paid off?

VFB Lingerie Europe began working with the Zetes voice system in September 2009. Now, having used it for only 6 months, the company has already recouped its investment and gained measurable benefits. Toni Solé, IT Manager of VFB Lingerie Europe, says: “In just six months, our initial expectations have been fulfilled. Zetes and its voice solution have enabled us to increase productivity by around 20%, and we can now offer a faster, more efficient service to various companies without any errors, unlike the previous paper-based system, where errors were inevitable due to the wide variety of products we handle.”

Now for those who are familiar with what I do for a living, it's important to note what this particular Zetes solution is, and what it is not. Using the terminology adopted by Wikipedia, this is speech recognition, not speaker recognition:

There is a difference between speaker recognition (recognizing who is speaking) and speech recognition (recognizing what is being said). These two terms are frequently confused, as is voice recognition. Voice recognition is combination of the two where it uses learned aspects of a speakers voice to determine what is being said - such a system cannot recognise speech from random speakers very accurately, but it can reach high accuracy for individual voices it has been trained with. In addition, there is a difference between the act of authentication (commonly referred to as speaker verification or speaker authentication) and identification.

So this particular Zetes solution can identify what is being said about lingerie, but cannot identify who is saying it. Cheating spouses can breathe easier, I guess.

P.S. It looks like I beat Ravi Das to this story!

P.P.S. FCC DISCLOSURE: My employer has done business with Zetes in the past. However, no lingerie was involved.
blog comments powered by Disqus