tts-error-reporting

To show multiprocessing TTS worker errors as frontend toast notifications, push structured error payloads from the worker through a shared response_queue, then intercept them in a response handler to trigger the UI toast. This ensures failures like quota exhaustion are surfaced immediately instead of stalling.

Multiprocessing TTS workers fail silently because separate processes cannot directly access the frontend UI. Without routing structured error payloads through a shared response_queue, failures like API issues or connection failures are swallowed and users remain unaware of why voice synthesis stalled.

The best way to propagate TTS quota exhaustion errors from a separate process is to define a standard error payload format and push it via a shared response_queue. A response handler intercepts these structured messages to trigger informative toast notifications on the frontend.

To format error messages from TTS workers for a response queue, push a well-formed error payload using a defined format like the __error__ tuple. This structured error propagation enables the frontend response handler to reliably translate multiprocessing worker failures into visible UI toasts.

Yes, this error reporting approach works for realtime TTS API connection failures. It ensures connection issues encountered by multiprocessing TTS workers are immediately surfaced as informative frontend toast notifications rather than crashing the worker silently.

Routing TTS errors through a multiprocessing queue requires all separate TTS worker processes to strictly follow the defined error payload format. If workers do not push structured error messages using the standard tuple, the response handler cannot intercept them to trigger UI toasts.