Implementing voice chat

[AblEdge]

To implement voice chat I transcript audio first then use generated text with chat Completion service to get text response.
according to models like gpt-4o-mini-audio is there any chat Completions to accept both audio and text?

[sophialagerkranspandey]

Yes there's a good sample here: https://devblogs.microsoft.com/semantic-kernel/using-openais-audio-preview-model-with-semantic-kernel/