Hacker News new | past | comments | ask | show | jobs | submit login

That would either increase the cost of the device or degrade the performance of the speech-to-text.

In the case of your phone, you'd get about an hour of battery life. Reasonably accurate speech recognition requires several orders of magnitude more computation than wake word detection.




That would either increase the cost of the device

On-chip speech-to-text has been around since the 80's. It's not expensive to implement, especially with modern manufacturing.

or degrade the performance of the speech-to-text

How accurate does it need to be? All the spy corporations need is a stream of keywords.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: