personaplex

The fact that qwen3-asr-swift bundles ASR, TTS, and PersonaPlex in one Swift package means you already have all the pieces.

Tricky thing is personaplex forked some of the kyutai code and has not integrated the LoRA stuff they added.

If anyone is curious it's a bit of a mess but it's on my GitHub under runvnc moshi-finetune and personaplex.

Then send the deepgram and personaplex text in as messages during the conversation and tell it to call a hangup() command when personaplex says goodbye or gets off track, otherwise just wait().

How do you prevent PersonaPlex from confidently answering something the reasoning model hasn't verified?

But my client saw the Sesame demo page, and so now I am trying to fine tune PersonaPlex.

The current S2S providers LiveKit offers include OpenAI, Gemini, and Grok and I'm sure they'll add Personaplex soon.

PersonaPlex is an impressive advance in full-duplex conversational AI, especially in how it combines natural turn-taking with flexible persona and voice control.

I have tried to take the fine tuning LoRA stuff from Kyutai/moshi-finetune and apply it to the personaplex code.

Does anyone have working code for fine-tuning PersonaPlex for outgoing calls?

I was so hyped about PersonaPlex from their demos, but in my test it was oddly dumb and unable to follow instructions.

I got PersonaPlex to run on my laptop (a beefy one) just by following the step by step instruction on their github repo.

PersonaPlex handles the "mouth" — low-latency backchanneling, natural turn-taking, filler responses at RTF 0.87.

Quoted from linked article: "PersonaPlex accepts a text system prompt that steers conversational behavior.