Getting delayed responses from the model and also some of the audio chunks are getting delayed, for example this sounds like, “The prime minister of India is… Narendra…Modi” even though the Transcript is coming on time. Can we know when will the fix for this come or multiple people are facing this issue. The model being used is Gemini-2.0-flash-live-001
Experiencing the same too. The initial chunks feel a break, in speech. I had a Rs:43,547 to be said, which due to the inconsistency has come out as forty-three rupees five-forty-seven. This is more evident in longer responses expected from the model.
This was all good, functional till last night. Something changed here?
It got fixed overnight but again today the same issue has happened, has something changed overnight again ?