Is it possible for the device to initiate TTS on its own, without user interaction?

[X-NoNAME]

Hi! First of all, I really admire your work—what you’ve built is truly impressive and inspiring!

I have a quick question about the project:

Is it possible for the device to automatically send a request to a TTS server, receive the audio response, and play it through the speaker?

I’d love to create a speaking clock—for example, one that announces the current time every 10 minutes.

I’ve already implemented a timer that triggers built-in sounds at the end, but I’m wondering how I could make it speak the actual current time instead.

Thanks so much for sharing this amazing project!

[78]

You should record the TTS generated audios first and convert them into OPUS OGG files. Place the files in the SD-card or assets partition and modify the code to call PlaySound() every 10 minutes.

[X-NoNAME]

Thank you for your quick response and for the suggestion!

However, I don't think the proposed method of pre-recording audio files will work for my specific use case. The goal is to announce the current time (e.g., "10:50 AM" or "23:20"), and it's not feasible to pre-record every single possible time combination (that would be 138 files!).

I thought about sending a "What time is it?" request to the server every 10 minutes. From what I've been able to figure out, sending it as text isn't possible. I'm now considering saving the UDP packets from a voice query and then sending those same packets every 10 minutes.