Issue changing default voice in Gemini2 live api

[notnotrishi]

Description of the bug:

I'm using the code in live_api_starter.py to test out the multimodal live api and build from there. The basic code works. Based on the documentation in https://ai.google.dev/api/multimodal-live#sessions I am attempting to change the default voice but it seems to be not working if I use the format mentioned in the API documentation:

My code snippet:

MODEL = "models/gemini-2.0-flash-exp"

MODE = args.mode

client = genai.Client(http_options={"api_version": "v1alpha"})

CONFIG = {
    "generation_config": {
        "response_modalities": ["AUDIO"],
        "speech_config": {
            "voice_config": {
                "prebuilt_voice_config": {
                    "voice_name": "Charon"
                }
            }
        },
    }
}

Error message:

Traceback (most recent call last): File "/Users/rishi/Projects/gemini_live/new.py", line 393, in asyncio.run(main.run()) File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/asyncio/runners.py", line 190, in run return runner.run(main) ^^^^^^^^^^^^^^^^ File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/asyncio/runners.py", line 118, in run return self._loop.run_until_complete(task) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/asyncio/base_events.py", line 653, in run_until_complete return future.result() ^^^^^^^^^^^^^^^ File "/Users/rishi/Projects/gemini_live/new.py", line 362, in run async with ( File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/contextlib.py", line 204, in aenter return await anext(self.gen) ^^^^^^^^^^^^^^^^^^^^^ File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/google/genai/live.py", line 626, in connect self._LiveSetup_to_mldev(model=transformed_model, config=config) File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/google/genai/live.py", line 459, in _LiveSetup_to_mldev _GenerateContentConfig_to_mldev( File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/google/genai/models.py", line 894, in _GenerateContentConfig_to_mldev t.t_speech_config(api_client, getv(from_object, ['speech_config'])), ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/google/genai/_transformers.py", line 301, in t_speech_config raise ValueError(f'Unsupported speechConfig type: {type(origin)}') ValueError: Unsupported speechConfig type: <class 'dict'>

Actual vs expected behavior:

No response

Any other information you'd like to share?

No response

[ThashilNaidoo]

I had the same issue. Try this instead. It worked for me.

"generation_config": { "response_modalities": ["AUDIO"], "speech_config": "Charon" }

[notnotrishi]

, "speech_config": "Charon"

I had the same issue. Try this instead. It worked for me.

"generation_config": { "response_modalities": ["AUDIO"], "speech_config": "Charon" }

that works, thanks @ThashilNaidoo !

but i also noticed the report is triaged as a bug so hopefully they can fix the issue and/or clarify which one is the correct usage

[MarkDaoust]

Hi, thanks for reporting this. I just looked into it and this is fixed in the latest release (0.4).

Both methods work now.

[brandonwheat]

Hi, thanks for reporting this. I just looked into it and this is fixed in the latest release (0.4).

Both methods work now.

I tried both and neither are working for me

Invalid JSON payload received. Unknown name "prebuilt_voice_config " at 'setup.generati; then sent 1007 (invalid frame payload data) Request trace id: 74133fd7e2c07425, Invalid JSON payload received. Unknown name "prebuilt_voice_config " at 'setup.generati

conn = await es.enter_async_context( connect( f'wss://{HOST}/ws/google.ai.generativelanguage.v1alpha.GenerativeService.BidiGenerateContent?key={API_KEY}') ) print('')

    initial_request = {
        'setup': {
            'model': MODEL,
            'system_instruction': {
                "parts": [
                    {
                        "text": SYSTEM_MESSAGE
                    }
                ]
        },
            "tools":  {'function_declarations': [pay_bill_tool, get_quote_tool]},
            "generation_config": {
                "response_modalities": ["AUDIO"],
                "speech_config": {
                      "voice_config": {
                        "prebuilt_voice_config ": {
                          "voice_name": "Puck"
                        }
                      }
                }
            }
        },
    }

[rubiagatra]

Let me try this

[Gunand3043]

Hi, I just checked on my end and it is working as expected. Please refer to the documentation, and let me know if you are still experiencing the issue.

Thanks

[github-actions[bot]]

Marking this issue as stale since it has been open for 14 days with no activity. This issue will be closed if no further activity occurs.

[github-actions[bot]]

This issue was closed because it has been inactive for 27 days. Please post a new issue if you need further assistance. Thanks!