What is a Voice Directed System?
A Voice Directed System (or Voice Automation System) is a combination of hardware
and software that enables employees responsible for certain type of job tasks to complete
their work faster and more accurately. The way such a system works is by providing the
employee the ability to receive instructions audibly and respond with observations or data
to be logged using a headset with a microphone on it. The Voice Directed System often
also offers additional user-directed functions. Examples are asking for more information
on a task, or informing the system of a discrepancy or special condition related to that
task.
What are the components of a Voice Automation System?
In most application the customer’s existing servers, databases, and software systems
often contain a great deal of information that is critical to the employee’s job function. To
get the information from the customer’s servers to the operator using a voice system, a
Voice Server computer is installed and standard wireless communication technology is
employed. Here are the hardware and software components of a Voice Automation
System:
Hardware
Radio Frequency (RF) Infrastructure – This communication network, often already in
place in many warehouses and manufacturing facilities, consists of wireless access points
connected to the customer’s existing network. This network typically employs the
802.11b/g, which are broadly known as WiFi.
Voice Server – This desktop or rack mounted computer server is the host of the main
voice application and the other administrative and reporting tools.
Voice Terminals – A voice terminal is an RF enabled hardware device that is worn or
carried by the voice operator, and that communicates to the Voice Server over the
wireless network.
Voice Headset – The operator wears a headset with earphones and a microphone which is
connected to the Voice Terminal. Connection to the Voice Terminal is either a wired
connection, or a wireless connection.
Software
Voice Application – The voice application consists of two general components:
 Off-the-shelf functions included in Vocollect’s software packages. Where applicable,
these functions, though embedded in the software, are configured by Speech Interface
Design, Inc. to meet the sites requirements.
 Custom programming providing extensions or additional functions as needed to meet
the needs of the customer’s facility.
The Voice Application has components that reside on the Voice Server and components
that are downloaded to the Voice Terminal.
VoiceLink – Off-the-shelf Vocollect Software “middleware” installed on the Voice
Server. It’s called middleware since it takes information from the customer’s existing
servers and databases, processes that information in accordance with the Voice
Application, and transmits it to the operator’s Voice Terminal. Conversely, it also
collects information from the operator’s Voice Terminal, primarily information spoken-in
by the user. VoiceLink also provides a rich set of reports that can be used by floor
supervisors to and site management to adjust how data is released to the operators and to
observe and analyze floor operations.
VoiceClient – VoiceClient is the Vocollect firmware that operates on the Voice Terminal,
and supports functions such as converting printed text to audible speech for the operator
to hear.
VoiceConsole – Vocollect’s administrative software that runs on the Voice Server
providing management of users, groups of users, system-access permissions, downloads
to the Voice Terminals, and related functions.
Can you describe in more detail how Voice Automation improves
employee productivity?
In warehouses and manufacturing facilities, certain job functions involve transactions that
are repeated over and over throughout the work day. Sometimes, these transactions
involve reading information from, or logging information to an instrument such as a
clipboard, sheet of paper, or an electronic device like a computer, PDA or barcode
scanner. Handling these instruments can add a significant amount of time to each
transaction; the employee must continually pick up and put down the instrument while
conducting the other parts of the transaction. Take the example of an order selector (or
“picker”) in a warehouse or distribution center. The picker may be filling an order
containing 32 items that are to be collected from various locations throughout a large
warehouse using a pallet jack. With a paper-based or hand-held scanner-based process,
the picker must read the next location from the paper or scanner screen, place the paper
or scanner on their pallet jack or in a holster, and take their pallet jack to the next
location. At the location they may once again refer to the paper or scanner screen to
verify the specific slot to pick from and the quantity to pick. After completing the “pick”
operation, they will repeat the process again until all 32 items are collected. So at each
warehouse location, the employee spends several seconds to a minute doing manual tasks
other than putting items on a pallet. Even if the extra time is only 15 seconds per item,
that means that whatever time it takes to fulfill that order includes 8 minutes of nonproductive
time. If that picker normally fills only 10 orders in a day (which would be a
very low number in most warehouses), 80 minutes of their day is consumed with
handling the instrument to read instructions. While this is one example, there are many
job functions in warehouses and some manufacturing facilities that have similar
characteristics.
A Voice Directed System, or Voice Automation System, makes it possible for the
employee (”operator”) to receive verbal instructions through a audio headset, and speak
observations or record data verbally into the headset. In a facility using Voice
Automation, each operator is equipped with a Radio Frequency based device (called an
RF “terminal”) that is often mounted on a special belt or on their wrist, and is connected
to the wired or wireless headset they are provided. The RF Terminal communicates with
a server which sends instructions to each operator and collects information on the task
they are conducting. In the warehouse example above, the picker is able to receive their
instructions, and ask for additional information while they are moving their pallet jack to
the next location. After confirming their location verbally, they can confirm the number
of items picked from that location, and start moving to the next location while listening to
information regarding their next location.
What sorts of job tasks are well-suited to benefit from Voice
Automation?

Voice Automation provides the greatest benefits when the job function being automated
includes one or more of the following characteristics:
• Hands-Busy Tasks–the user’s hands are occupied while data entry still needs to be
performed.
• Eyes-Busy Tasks–the user needs to be looking at a part or object while entering
data, such as inspection of parts.
• Data Entry is a significant portion of a user’s job.
• Real-Time Data is required for input to a Warehouse Management System, an
SPC Charting program, to print a “load slip” or to confirm a materials receipt.
• Two-Person “Tally” applications–a second “tally” person is required to record
data the inspector collects.
• Non-Bar Coded Data needs to be entered, possibly for conversion to a bar code
label.
• Attribute or Qualitative Data is involved, such as “broken”, “scrap core,”
“missing component,” “glaze defect”, etc. This data can be spoken-in by name,
vs. the inspector needing to remember codes.
• More specific data is needed, but which cannot be entered with the current data
collection methods.
• The Data to be entered is relatively standard and repeatable: part numbers, serial
numbers, quantities, attribute defects, locations, etc.
What are some examples of Voice Applications that have been
developed in the past?

• Warehouse Applications such as Order Selection, Replenishment, Put-Away,
Cycle Counting, Cross-Docking and others.
• Receiving and Receiving Inspection
• Automotive Paint/Final Inspection of painted chassis or fully assembled vehicles
• Ceramics Inspection Systems for final inspection and disposition of ceramics after
firing
• Automotive Cores Receiving for the Automotive Parts Remanufacturing industry
• Electronic PC Board inspection at an inspection “bench” and especially for use
with a microscope
• Voice to Bar Code Label situations for conversion to a machine readable format
• Voice-Validated Serial Number Entry, where technology and good process
provide accuracy
• Package Identification for material Handling, such as postal sorting