[sim04ful]

The main issue I'm facing with realtime responses (speech output) is how to separate non-diegetic outputs (e.g thinking, structured outputs) from outputs meant to be heard by the end user.

I'm curious how anyone has solved this

[artur44]

A simple way is to split the model’s output stream before TTS. Reasoning/structured tokens go into one bucket, actual user-facing text into another. Only the second bucket is synthesized. Most thinking out loud issues come from feeding the whole stream directly into audio.

[pugio]

There is no TTS here. It's a native audio output model which outputs audio tokens directly. (At least, that's how the other real-time models work. Maybe I've misunderstood the Qwen-Omni architecture.)

[artur44]

True, but even with native audio-token models you still need to split the model’s output channels. Reasoning/internal tokens shouldn't go into the audio stream only user-facing content should be emitted as audio. The principle is the same, whether the last step is TTS or audio token generation.

[regularfry]

There's an assumption there that the audio stream contains an equivalent of the <think>/</think> tokens. Every reason to think it should, but without seeing the tokeniser config it's a bit of a guess.