ProMat is the largest materials-handling and supply chain conference in North America. It meets every year in Chicago; the exhibit hall is filled with conveyors, cranes, pallets and other warehousing equipment; consultants; and various types of software systems. Interspersed among the other software vendors are speech-technology companies, notably Vocollect, Voxware, Lucas, and TopVOX. This is nothing new; speech vendors have held their place in the ProMat exhibit hall for many years. but this is the first year that other exhibitors have called speech “sexy.”

In the past, companies with warehouse management systems (WMS), handheld devices, and other offerings have viewed speech primarily as a secondary option they were considering or were adding to their product line. That is, until around two years ago, when speech suddenly became a technology of interest for their customers. For the first time, I heard ProMat attendees asking each other about speech as they walked the aisles of the exhibit hall. A vendor of a competing technology complained that customers were so interested in speech they didn’t want to hear about anything else. She went on to accuse me of contributing to the speech-technology frenzy with this article. I pled guilty.

There are a number of reasons behind the emergence of speech. One is that the number of happy speech customers has reached a tipping point. Their successes demonstrate that speech works in noisy environments when used by warehouse workers and that the return on investment can be seen within a year.

The growth of industry partnerships has also enhanced the allure of speech. Speech companies support widely used handheld devices manufactured by Motorola/Symbol, Psion, LXE, Intermec, and others. These partners now have devices that are audio- and voice-capable and some, such as Motorola/Symbol and Intermec, have extended their certification programs to include speech. WMS partners, such as Red Prairie, are playing a central role in the popularity of speech, as well. When speech is integrated into a WMS, it uses the WMS’s logic and resources, which provides an attractive consistency and ease of deployment. 

Picking is typically the first speech deployment. It involves retrieving and assembling the items ordered by a customer. The system proceeds through an order, item by item, telling the worker where to find each item and how many to get (pick) for the order. Pickers can also report problems (e.g., damaged goods, the stock needs replenishment).

When asked why they use speech, many with whom I spoke mentioned that they were attracted by the promise of increased productivity and the quick realization of an ROI. I also suspect that the recent economic downturn has further heightened interest in technologies that keep expenditures low while boosting productivity. Customers also cited the hands-free aspect, support for multiple languages, and the ability to work with other technologies supported by the same handheld or wearable device. One TopVOX customer turned to speech to increase productivity and enhance worker safety. He began using speech for the picking of small items and then expanded its use to case picking and then all picking. More than a year ago, he added replenishment and is now working on put-away.

Each of the vendors with whom I spoke has taken a different approach, both in terms of technology and markets. Pittsburgh-based Vocollect has ported its technology to standard devices, but reports that its own wearable device (TalkMan) remains the most popular platform. The company’s newest release is the T5 which, among other things, is Bluetooth-capable. Vocollect lays claim to being the market leader in the U.S. and has deployments in Europe and Asia. 

Voxware, which is based in New Jersey, is the oldest speech company serving the warehousing industry. In the past few years, Voxware has intensified its focus on software solutions using a services-oriented architecture (SOA) rather than hardware platforms and is one of the rare companies using VoiceXML for non-telephony applications. Although Voxware has deployments that are outside of North America, its market focus remains on North America.

TopVOX is a subsidiary of the German company Top. It offers speaker-independent speech as well as speaker-dependent technology. An important advantage of speaker-independent technology is that it eliminates the need for new workers to “train” the system to their voices and to create an individual voice model. The challenge is that a speaker-independent model must be developed for each language and each major dialect. TopVOX has models for the major languages of Europe and is developing models for a number of other languages. TopVOX has global operations and is the market leader in Europe. 

Lucas Systems, based in Wexford, Pa., sells speech solutions using speech technology from an (unnamed) ASR vendor.  The company uses an SOA  approach based on .NET and XML coding, but not VoiceXML. It is fully integrated with standard handheld devices, such as Motorola/Symbol and Intermec, but the company offers its own headsets. Lucas Systems’ market is North America only. 

Judith Markowitz, Ph.D., is president of J. Markowitz Consultants and a leading independent analyst in the speech and voice biometrics fields. She can be reached at judith@jmarkowitz.com.