The core value of voice is freeing the hands and eyes
Aliases: hands-free · eyes-free · occupied hands · occupied vision
What it is
Voice as input is not valuable because speaking is cooler than typing. It is valuable when the hands are occupied and the eyes cannot leave the environment, yet a command still has to be issued. That is hands-free and eyes-free. Kneading dough, carrying bags while walking, holding a wheel: under those constraints, voice has a place touch and keyboards cannot take. Dictating at a desk toward a screen is mostly speed or preference, not liberation.
Why it happens
Touch and key-and-mouse need an effector on a control surface and usually a visual loop to aim. When both hands hold tools, or vision must watch a road or a pan, that loop breaks. Voice swaps the effector for the vocal tract and feedback for audition or a glance, so the body can keep its posture. Value rises with constraint hardness: the harder the constraint, the more recognition error is worth paying. Conversely, when the scene occupies neither hands nor eyes, voice still costs utterance planning, waiting, and correction, and the trade falls apart. Hands-free and eyes-free are often named together and can split: headphone dictation can be eyes-free while still confirming with a gesture; a phone in a pocket can be hands-free for a call and still demand a look to tap agree.
Studying it
Use dual tasks: a primary task that occupies hands or eyes (tracking, a driving simulator, assembly) and a secondary task done by voice or touch. Dependent measures include primary-task degradation, secondary completion time, eyes-off-road time, and workload. The comparison is channel choice under constraint, not dictation speed in a quiet lab. If participants are allowed to pause the primary task and look, the eyes-free condition is broken; check with gaze or steering data.
Where it stops holding
Some disabled users rely on voice even when hands and eyes are free; the value is access, not situational occupancy. The social cost of speaking in public can outweigh hands-free gain. When recognition is poor, liberation turns into repeated respeaking and the hand goes to the screen anyway. Labelling every voice feature hands-free hides the many sessions that are in fact used while looking.
Applying it
- List the real situations where hands or eyes are occupied, and make voice the primary path only for those commands.
- Do not accept desktop dictation against a hands-free bar; accept driving and kitchen packs with both hands staying on the primary task.
- Check with dual tasks and gaze: while the voice command completes, are the hands and eyes still on the primary task.