Requesting support for user voice interrupts when using the Gemini Live API. Currently, the model keeps talking (often for 10+ seconds) about an old topic even when the user speaks over it (interrupting it).
Relevant documentation section: https://firebase.google.com/docs/ai-logic/live-api?api=dev#handle-interruptions
Looks like the Python library already supports this. Does this mean only client SDK changes are neccesary?
Requesting support for user voice interrupts when using the Gemini Live API. Currently, the model keeps talking (often for 10+ seconds) about an old topic even when the user speaks over it (interrupting it).
Relevant documentation section: https://firebase.google.com/docs/ai-logic/live-api?api=dev#handle-interruptions
Looks like the Python library already supports this. Does this mean only client SDK changes are neccesary?