Support speech recognition and text to speech

[lynchee-owo]

No description provided.

[mondaychen]

You are using useEffect in the PR. It's probably the most misused hook of React. Although this PR is still in draft, I want to remind you to think about the lifecycle of the effects, and how to make sure the dependency array is carefully set up.

If this is new to you, please spend some time read

• https://react.dev/reference/react/useEffect#specifying-reactive-dependencies
• https://react.dev/learn/lifecycle-of-reactive-effects

[lynchee-owo]

You are using useEffect in the PR. It's probably the most misused hook of React. Although this PR is still in draft, I want to remind you to think about the lifecycle of the effects, and how to make sure the dependency array is carefully set up.

If this is new to you, please spend some time read

• https://react.dev/reference/react/useEffect#specifying-reactive-dependencies
• https://react.dev/learn/lifecycle-of-reactive-effects

I think for handling the case "run task immediately" when stop listening, could use an alternative approach than creating an event "document.dispatchEvent(new CustomEvent("stopListening"))" and listen it, could rewrite it in the state management way

[mondaychen]

I think we should move this to the settings state since it should be persisted. But if you are working on the settings view as your next task, we can take care of that as well.

[lynchee-owo]

good point! I will take care of this during the settings view implementation.