Voice command has a number of advantages over a conventional keyboard, or touch screen, input. Perhaps most importantly, it can be used when the operator’s hands and vision are engaged in another activity. For example, using voice commands in a vehicle, a driver can control navigation, heating, music and a phone, all without taking their hands off the wheel or eyes off the road. Another advantage of voice is that complex instructions can be given much more rapidly than even a skilled typist could enter them into a keyboard.
Voice command is often combined with Natural Language Processing (NLP), which enables a computer to understand the meaning of sentences rather than simply recognising standard commands. This means that, in theory, complex instructions can be given as a natural conversation between the operator and the controller. The operator should be able to tell the controller what needs to be done; the controller can respond by asking for further clarification where anything isn’t clear and then ask for final confirmation before carrying out the operation. In practice, NLP is not fully mature, and results can be mixed.
An example from industry, where simple voice commands can keep the operator’s hands and vision free for other tasks, is in large scale measurement. Laser trackers are used to make high-accuracy coordinate measurements when manufacturing products, such as ships, aircraft and wind turbines. These instruments use a laser to track a reflective target over large structures. An operator measures features by probing them with the reflector while the laser tracks the reflector. Without voice command, a second operator is often required to press the measure button and check that valid measurements have been made. With voice command, a single operator is able to both position the reflector and operate the instrument, issuing commands such as “measure” and “delete last measurement”. Similar advantages may be gained in machine set-up or assembly tasks.
EFFICIENCY BOOST
One of the first machine tool manufacturers to use voice was Mazak with its Voice Advisor system, first introduced in 2007. This system simply used text-to-speech to give feedback. This means the machine used a synthesised voice to issue notifications rather than just displaying them on the screen, for example, to inform the operator which axis had been selected. Mazak’s original Voice Advisor did not accept voice commands.
Iconics Voice Machine Interface (VMI) works with voice-based digital assistants, such as Microsoft Cortana, Amazon Alexa, and Google Assistant. It interprets voice commands to perform various functions in manufacturing, energy management, industrial automation, or smart buildings. It can be used to monitor the status of systems and processes, control equipment and devices, and analyse KPIs. This is said to provide the convenience of hands-free natural language interaction while boosting productivity and operational efficiency. Iconics has integrated consumer voice command into industrial systems in the past. It used Microsoft Kinect for speech recognition, as well as tracking body movements to navigate through 2D and 3D graphics, as well as to issue commands.
The key advantage of the new generation of voice-based digital assistants is that they can interpret natural language rather than simply recognising standard command words. A plant manager might say, for example, “Hey Cortana. What is the OEE for Line 1, and how does it compare to last shift?”.
Other examples given for how Iconics VMI might be used include a facility maintenance engineer and a company executive. “Alexa. What is the current voltage of this chiller compressor?” asks the maintenance engineer, to which the VMI responds, “The current voltage of this chiller compressor is 200V at 50 Hz”. The executive asks, “Ok Google. What is out total oil production year-to-date, and how does it compare to last year?” The VMI answers this query with, “You have produced 5,640,000 barrels of oil year-to-date. That is up 4% from last year.”
UNIVERSAL INTERFACE
In perhaps the most ambitious application of voice command technology, ITSpeeX have produced the Athena voice operated assistant for machine tools. This is intended to provide a universal interface that will work with any controller, allowing operators to easily move between different machines without learning their way around the different controllers. The Athena system doesn’t replace a machine’s controller but rather sits between the controller and the operator, effectively replacing the keyboard and menu system.
It can provide reports on machine status, control the machine, answer questions such as, “How do I change the coolant”, and perform calculations, such as process capability. Hundreds of functions can be performed by issuing simple spoken commands. Athena has been integrated into the controllers of machine tools from DMG Mori, Makino and OKK.
As an example of how complicated menu systems can be replaced with simple commands, the interaction for a tool change would be something like the following:
● “Athena. Tool change, tool five.”
● “Are you sure you want to change to tool five?”
● “Athena. Yes.”
● “Changing tool five.”
Mark Waymouth, Athena project manager at Makino, argues: “This dramatically simplifies the operation of the machine. The easiest human-machine interface we could think of is speech… We’re looking at Athena to be able to help set-up, run, utilise the tooling, maintain the machine, and learn the machine… One of the things Athena does is help coaching. What’s one of the easiest ways initially to coach? Providing access to manuals.”
With typical manual systems it can take a long time to find the instructions you need. With a voice assistant it can be as simple as asking, “Athena. How do I change the air filter?” to which you would hear the response, “Air filter change instructions displayed” as they appear on the screen.
“All of a sudden, 20 minutes that would have been wasted is there right now. There’s so much data noise out there. Athena’s going to provide the pathway to interface with the people,” Waymouth adds.
Voice command looks set to revolutionise the way we interact with machines and data. Speech recognition allows hands-free and eyes-forward interaction. It becomes possible to issue a command or request information while still actively engaged in another task. Combine this with NLP and it’s no longer necessary to remember how to navigate complex menu systems or to memorise many command words. You can simply ask for what you want. This is already enabling the operation of complex machine tools without lengthy training. In the near future, expect to see further improvements in collaborative robotics as humans and robots start to work closely together within the same work spaces.