I'm curious if speech endpointing by keyword is possible / will be supported in the near future? For example, depending on what question gets asked, I might spend more time thinking through a response or start speaking only to pause and reconsider my thoughts for longer. Would love to be able to override the more active endpointing that currently gets done.

@eric Absolutely, speech endpointing by keyword is an interesting concept and can be a valuable feature for voice-driven applications. It allows for more dynamic interactions and can accommodate different speech patterns and thought processes. While I can't speak to our specific roadmap or future feature releases, I can say that incorporating user feedback and exploring new capabilities like keyword-based endpointing is always on our radar. If you have a particular use case or further ideas on how you'd like to see this implemented, please feel free to share more details. Your input is invaluable in helping us improve our services and consider new features that meet the needs of our users. If you need more technical details or want to discuss this further, please ask support for more info on this.

Can you provide some more conext about it. It might be useful for you @ruMblE

We already support custom endpointing rules based on regular expressions.