Chat UIs with native audio input for multimodal models?

I've been running Gemma 4 E4B with oMLX and I can't find any chat interfaces that directly send the audio file to the model instead of running the audio through a separate STT layer. I can confirm the audio layers work because I ran a couple of requests through Pydantic AI in the Python REPL.

Thanks in advance.

EDIT: I know that llama-server's web UI can do this, but I don't feel like running an instance of llama-cpp just for the UI.

EDIT2: Reason why I am asking is because I want to try using Gemma 4 as a lower-latency voice assistant.