Pick-by-Voice

How does pick-by-voice work?

Pick-by-voice works on the dialogue principle. The higher-level system – a WMS or the warehouse module of an ERP – provides the picking orders and passes them to the voice terminal. This converts the data into spoken prompts via speech synthesis (text-to-speech): the worker first hears the storage location to head for. On arriving at the rack, they read aloud a check digit displayed there. Only when this matches the expected digit does the system release the next step and announce the pick quantity.

After the pick, the worker confirms the quantity by voice. Speech recognition (speech-to-text) translates the answer back into a data record that is reported to the WMS or ERP immediately. This creates a closed loop of prompt, action and confirmation that guides the worker step by step through the entire order. Because every line item is confirmed individually, the processing status is transparent at all times.

Why pick-by-voice? Benefits and relevance

The key advantage of pick-by-voice lies in hands-free working. Because the worker has to hold neither a paper list nor look at a display, they can use both hands to grab items and concentrate on the route. This noticeably increases the pick rate and improves ergonomics at the same time. Studies and field reports from intralogistics regularly cite productivity gains in the double-digit percentage range compared with paper-based picking.

The second plus point is quality: the mandatory confirmation via check digit significantly lowers the error rate, because mix-ups of storage locations are caught early. New employees are also quickly trained in, since they simply follow the prompts – language barriers can be cushioned with multilingual prompts. The paperless way of working also saves paper and printing costs and provides the ERP with stock levels reported back in real time.

Pick-by-voice in the ERP and WMS context

Pick-by-voice is not an isolated system but a front end for picking that draws its data from the leading system. In smaller operations the ERP itself often handles warehouse management and provides the pick tasks; in larger warehouses a specialised WMS sits in between, optimising routes, bundling batch picks and driving the voice control. What matters is a clean interface through which orders flow out and confirmations flow back.

For operations this means: item master data, storage locations and stock levels must be maintained in the ERP so that the prompts are correct. The feedback from the voice dialogue updates inventory management and order status immediately, so that goods issue and the delivery note can follow on directly. Many mid-market ERP and multichannel systems offer voice connections either natively or via certified logistics partners.

Distinction: pick-by-voice vs. pick-by-light and other methods

Pick-by-voice is one of several paperless picking strategies that differ in their guidance technology. It is worth comparing it with the common alternatives, because no method is universally superior – the choice depends on the assortment, throughput and warehouse structure.

Typical use cases and limits

Pick-by-voice plays to its strengths wherever many line items are picked from rack warehouses and hands need to be free: in food wholesale, in beverage logistics, in deep-freeze warehouses (thick gloves instead of a touch display) and in spare-parts and pharmaceutical distribution. The method is also widespread in e-commerce fulfilment with high shipment volumes.

Limits show up in very noisy environments where speech recognition suffers, as well as in tasks that strictly require visual inspection (such as quality control by appearance). For very small warehouses with few daily picks the investment often does not pay off – here a paper or scanner solution remains more economical. An honest cost-benefit assessment before implementation is therefore indispensable.