Started two days ago. Happens both on the API and the Google AI Studio. I’m experimenting with Turkish language specifically, doesn’t occur for English.
Hi @doneforaiur ,
Could you please share a minimal reproducible example including the exact prompt you’re using, the selected Turkish voice, a sample of the generated audio and the specific model you’re using (gemini-2.5-flash-preview-tts or gemini-2.5-pro-preview-tts) along with the error screenshots? This will help us reproduce the degradation and artifacts issue on our end.