AI-generated Answer for itsdun6402
If you're not receiving the isFinal=True flag from the TTS WebSocket, it might be due to how the end-of-stream is being signaled. To ensure you receive the isFinal=True response, you should send an End Of Sequence (EOS) message with an empty string ({"text": ""}) instead of a flush message with a space. This signals the server that the text input is complete, and it should return a final response with isFinal=True and no audio data.
If you want to maintain a long-lived session, you may need to re-establish a new connection for each text segment or manage the session by using flush messages and accumulating alignment data without relying on the isFinal flag. This approach ensures that each text segment is treated as its own task, and you receive the final signal appropriately.
rotund badge
nimble crane