Skip to content

Gemini Live API is returning multiple chunks of answers, OR restarting answers for no reason, for a single input #952

@BatmanofZuhandArrgh

Description

@BatmanofZuhandArrgh

Description of the bug:

The code is quite close to the sample code.

For a single input, it keeps stop vocalizing, triggering turn = self.session.receive(), then restart the answer all over again. This only happens when the mode is "AUDIO", not "TEXT"

For example, results for 1 input audio is this. All for 1 input.
========== New turn ========== (Print after self.session.receive()) Received audio data: 9600 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 7680 bytes Received audio data: 9600 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 3840 bytes Received audio data: 9600 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes ========== New turn ========== Received audio data: 9600 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes ========== New turn ========== Received audio data: 9600 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 11520 bytes Received audio data: 1920 bytes ========== New turn ==========

THE CODE:

audioloop.py

Please help me, if I'm missing something

Actual vs expected behavior:

No response

Any other information you'd like to share?

No response

Metadata

Metadata

Assignees

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions