If Amazon or Alphabet have their way, “Alexa” and “OK Google” may be the most common spoken words in US households by 2018.
Similarly, as other devices take the stage for ambient voice interaction, it’s likely they’ll offer up their own wake words (e.g. “Hey Siri” or “Hello Bixby”).
Discussed in the past was how false trigger or false rejection was one of the most frustrating experiences with ambient voice interactive devices. However, there are some ways to improve interaction through dynamically changing the thresholds for rejection of triggers. Some possibilities include:
- Increasing sensitivity when audio is playing, especially for commands like “stop” or “pause”
- Decreasing sensitivity when ambient noise is detected, especially conversation
- Increasing sensitivity for particular phrases or commands depending on their anticipation use
Other ways to make false trigger friendlier are to duck audio during music play for the receipt of a new command without playing a loud tone or showing a visual indication or have a visual only indicator when ambient noise is low.