Frequently Asked Questions
While each voice implementation is unique, our customers have seen on average productivity improvements from 7% to 50%+ for each application, within the first year of implementation.
Being able to actually hear the error spoken instead of just a beep eliminates a large percentage of errors. Each implementation is unique. Significantly reducing manual entry errors (fat fingers) and automating keystrokes can deliver double digit improvements in accuracy alone.
If you looked at voice solutions that were server based and required a significant capital outlay to implement, then you looked at a legacy voice solution that may have been created up to 20 years ago. The AccuSpeechMobile voice solution architecture is to deploy our application on the mobile clients, without needing to make any changes to the WMS server you have in place today.
We provide an ROI calculator that assists with quantifying an investment in voice enablement.
Every installation is different, but the majority of our customers go live in 90 to 120 days.
No. AccuSpeechMobile is user independent and can be used immediately by your existing team and is easy to learn for new or temporary team members.
Our technology works in numerous noisy environments, both inside and outside the warehouse. As we roll-out the solution, we customize the implementation to take advantage of every capability to ensure outstanding voice recognition.
No. AccuSpeechMobile is Speaker Independent, modern voice recognition technology that is designed to work well with a variety of voice input situations.
Many server based voice picking solutions require user training and they keep user profiles on the server. These are speaker dependent solutions. With a speaker dependent solution a custom configuration is needed for each user. Each user must create a voice profile and each profile must be saved and stored for use, loaded on a users device and then changed for each new person using the device. With a speaker independent solution, one configuration is used for all users. A device based solution that is speaker independent, combined with today’s state-of-the art speech recognition technology, eliminates the need for the older speaker dependent model. AccuSpeechMobile is speaker independent, with no server requirement or hours long training of the system to properly recognize a particular voice.
We currently support over fifty languages and have installations around the world.
Voice capable devices (has a audio codex with audio in and out capabilities via wired headsets or Bluetooth 4.0 and HFP 1.6 support) with Windows, CE5, CE6 , CE7 and Android (4.1 or newer) operating systems. We recommend using a headset for all voice related applications.
No, AccuSpeechMobile is client based solution that works entirely on the mobile device.
None. Everything is loaded on the device.
AccuSpeechMobile can be integrated with an in-house developed WMS system as well as popular Warehouse Management Systems such as IBM DView, Manhattan WMS, HighJump WMS, LogPro WMS, Oracle WMS, Red Prairie WMS, Robocom WMS, SAP and more.
Yes. We recommend a review of the network to ensure there are no dead spots that might interfere with communications between the handheld and the WMS host.
No. AccuSpeechMobile is a client based solution which translates the voice command into digital data at the device. Data from the device utilizes the existing network in exactly the same manner as before installing our solution.
No special requirements are needed. The network needs to support the bandwidth and the number of devices used and should support the applications needs with minimal lag time to update the information on the screen. The network should provide adequate coverage to ensure there are no Wi-Fi signal dead spots.
Voice is not just for picking anymore. With the market acceptance of device based voice, using voice for many warehouse applications has become commonplace. Voice enablement is suitable for picking (all types) as well as applications such as cycle count, replenishment, shipping, receiving, break pack, cross docking as well as maintenance and field services applications.
With AccuSpeechMobile, workflow automation can be customize for your particular application (on the client) eliminates manual user interaction on the mobile device and reduces user looks at the screens and hence impacts worker productivity significantly.
Voice interfaces can be used for many warehouse applications. While voice picking was the first and most logical application, many companies are finding additional applications that provide rapid, measurable ROI including shipping, receiving, cycle count, returns, cross docking, and others.
Any application that can benefit from freeing the user’s hands, reducing the interactions necessary by the user and reducing the need for users to visually look at screens to continue workflow could be a candidate for voice enabling. Applications that can be improved through automation in addition to reduced user interactions and looks will benefit from adding voice.
Yes. There are customers with legacy voice picking (Vocollect, Voxware, Lucas) that have installed AccuSpeechMobile for other applications, or completely replaced the older server based applications with the AccuSpeechMobile device based, serverless approach.
Voice capable devices (has a mic and can process audio input) with Windows, CE5, CE6 , CE7 and Android (4.1 or newer) operating systems. We recommend using a headset.
Devices are not user dependent. Many of our customers run multiple shifts with users sharing devices. That said, we highly recommend each user has their own headset.
Maintenance for existing devices is the same as prior to installing AccuSpeechMobile; no additional maintenance is required. A typical customer will have an RF device and scanner combination, and add a headset. Most customer’s prefer to assign a headset to each user for hygiene purposes. Maintenance of headset may include replacement of the foam ear piece, foam wind cover over the MIC and connector cables (if using a wired headset).
Yes, devices with Android version 4.1 or higher.
We have a network of experienced partners and relationships with handheld scanning leaders such as Zebra and Honeywell among others. If you are still picking off paper you would need to deploy the mobile devices and then add our voice solution. Please ask us for partner referral information.
After the first few days of getting used to a voice system, it becomes second nature. Since the application that has been voice enabled is the same one that has been used in the past without voice, the key to a successful implementation is to ensure the users know what is supposed to be said in each input filed and screen. We recommend laminated cards be printed up for the first month. They can then be used to quickly train new and seasonal employees as well.
A new user may take a few minutes to a few days to learn all the voice commands. Certain application screens may only be needed for special situations. As the user becomes comfortable using a voice directed system, performance will typically continue to improve for several weeks.
The type of feedback can be customized to your work requirements. There are many effective methods, for instance, we can provide audio feedback: a beep/tone, or speaking back a work such as ‘completed’ or ‘continue’.
After the customized voice interface is built, the AccuSpeechMobile Runtime license is loaded on the handheld device. Next, two small files are loaded from the AccuSpeechMobile Studio and connected to the AccuSpeechMobile Runtime which enables a fully functional, voice enabled device.