Getting Partial Response from gpt4o realtime model

Community
,
1

image
image1737×353 52.1 KB

Getting partial response from the model, can someone please guide further.

Model - gpt4o
endpoint - realtime

I am working on voice to voice functionality.

Any help will be appreciated.

Thank you

2

Thanks everyone, I was able to figure out a solution. Changing the max_output_tokens value from 1024 to 4096 helped. Sample code below.

    await ws.send(
        json.dumps(
            {
                "type": "response.create",
                "response": {
                    "modalities": ["text", "audio"],
                    "instructions": """talk in english, respond to the user query, and ensure you complete your entire response""",
                    "voice": "alloy",
                    "output_audio_format": "pcm16",
                    "temperature": 0.8,
                    "max_output_tokens": 4096, # updated this from 1024 -> 4096
                },
            }
        )
    )