UserAgent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.77 Safari/537.36

Steps to reproduce the problem:
I have a web application that utilizes the Speech Web Api for speech recognition that allows the user to issue commands, some of which navigate to other pages in the application (same domain) or reload the page. When the page is navigated or reloaded, the user cannot resume interaction with the application using their voice because the audio feedback mechanisms (web audio api, speech synthesis api currently deprecated, but will be the same problem as web audio api) cannot be used until the user has interacted with the page using a physical device (mouse, keyboard).

What is the expected behavior?
The onresult event on the SpeechRecognition object should count as user interaction that satisfies the new autoplay policies. The user has already been required to give permission to access the microphone, after which the microphone should be treated as any other input device that allows interaction with the page.

What went wrong?
Feedback sounds, and soon synthesized speech, in response to user microphone input via SpeechRecognition are prevented from playing until the user interacts with the page using a device other than the microphone. Handicapped users are forced to interact with the application using a physical device.

Did this work before? N/A 

Does this work in other browsers? Yes

Chrome version: 70.0.3538.77  Channel: stable
OS Version: 10.0
Flash Version: 

It seems allowing the microphone to satisfy the autoplay policies does not open a security risk or potential for abuse because the user has already granted the site permission to use the microphone as an input device.