Hello, we are experiencing problems with the Live sessions we implemented in our app. We’re using model ‘gemini-2.5-flash-native-audio-preview-12-2025’ but it doesn’t seem to initiate the conversation, and only responds whenever there is input from the user. We would like the AI voice to greet the user first, before responding to input of the user. We have tried multiple Gemini AI models, but apparently, the 2.5 flash native audio should be the best one. However, we don’t seem to get it to work… Is there anyone who can help us with this? We’d be forever grateful!!
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Gemini 2.5 Flash Live API native audio -- can't get it to initiate the conversation | 6 | 285 | November 25, 2025 | |
| Can't make the model start speaking with Gemini Live and the latest model | 1 | 132 | December 29, 2025 | |
| Gemini Live API experiencing audio glitches | 1 | 100 | February 20, 2026 | |
| Gemini Live Audio WebSocket closes immediately after opening | 3 | 67 | March 29, 2026 | |
| Suddenly the Gemini Live API stopped understanding input audio | 6 | 689 | November 12, 2025 |