Voice assistants should listen to Nike a little more and Just Do It. While Google Assistant / Alexa are general better at this, others can bring a lot of pomp to their performance of tasks.
- Me to Bixby: Turn on hotspot
- Bixby: Alrighty, I’ll do that.
It takes me to the menu to turn on hotspot (why this is no longer available on the swipe down tray is something else) but if I’m already seeing you take me there, I don’t need to the TTS response. This is especially if I’m doing this in a coffee shop or some place where its still awkward to have an out loud conversation with my phone. This could be acceptable if headphones were connected to the device but not necessarily while the device or myself are stationary and ambient noise low.
Sometimes a voice response is not needed. Typically, this is:
- When there’s a display that shows the action
- When an audio response will show the result has been performed (e.g. playing music)
- When there’s a physical actuation that confirms the action (e.g. light turns on)
Normally, we’re instinctively able to vary the voice response to asks of us. Voice first devices can adjust the response from no response, to terse, to full depending on:
- Ambient noise level
- Time of day
- Presence of more than one individual
- Location of the user (both proximity to device as well as in / out of the home)
The aim is to make a voice interactive device into a more considerate intelligent being.