Speech Recognition in the Warehouse – Voice Picking

Although many consultants, trade magazines and marketing literature use the terms 'voice recognition' and 'speech recognition' interchangeably there is a difference. Voice recognition is a method of capturing the voice to identify the person that is speaking. This would be similar to identifying someone by his or her fingerprint. Speech recognition, by contrast, is the method of capturing the voice to identify and to return what someone has spoken.

Speech recognition is accomplished by breaking the captured human voice into text consisting of words or phrases. For example, in a mobile order picking application for a warehouse, when an operator is directed to move five cases of product ABC, the product is moved then the operator says a completion phrase like, “Got It”. Next, the computer captures the phrase and returns the text “GOT IT” to the calling program, which then prompts the database to reduce the number of product ABC by a quantity of five.

Speech recognition systems are used primarily for order picking from a staging area to a pick belt or conveyor and from racks to carts or totes. Fork lift operators are beginning to use the technology as well now that Bluetooth wireless headsets are improving in speech quality. The use of speech recognition systems such as offered by Datria, VoCollect, Lucasware, and Voxware are the most prevalent.

Speech recognition is less reliable when an operator has to speak a long sequence of alphanumeric characters such as serial numbers or UPC codes. This type of activity is best left to RFID and barcode scanning.

What tends to hold companies back from implementing a voice technology system?

The myth that deploying a voice-based solution will disrupt warehouse activities during the development, pilot and migration to voice technology can hold managers back from implementing a system. No time is ever the right time to slow down the order fulfillment process that requires voice technology.

In truth, implementing voice technology is no more intrusive than performing normal business operations as holding site tours, upgrading software and training the team. Adding speech recognition and voice guided work instruction to a facility tends to bring excitement of something new to the workers and invariably job satisfaction is improved.