how do I pass output_format when creating a conversational ai agent

[rancid belfry]

AI-generated Answer for hshreeshail

Currently, there is no mention of passing output_format directly when creating a conversational AI agent. However, you can override the audio input/output type by patching the agent settings before each conversation begins. For more details on how to update these settings, refer to the API documentation.

If you need to standardize audio output for text-to-speech, you can explicitly set the output_format parameter when initiating a websocket connection or using the streaming endpoint. For example:

• Websocket connection:
  wss://api.elevenlabs.io/v1/text-to-speech/{voice_id}/stream-input?model_id={model}&output_format=mp3_44100

• Streaming endpoint:
  https://api.elevenlabs.io/v1/text-to-speech/{VOICE_ID}/stream

Let me know if you need further clarification!

[nocturne rover]

so, I cannot set the format of the audio output in a conversational agent?

[rancid belfry]

AI-generated Answer for hshreeshail

Currently, the only way to override the audio input/output type for a conversational AI agent is by patching the agent settings before each conversation begins. Unfortunately, there isn't a method to dynamically set the audio format at runtime during the conversation. For more details on how to update these settings, you can refer to the API documentation.

[nocturne rover]

how to patch the settings before the conversation begins?