NVIDIA Releases PersonaPlex-7B-v1: A Real-Time Speech-to-Speech Model Designed for Natural and Full-Duplex Conversations
By Asif Razzaq
NVIDIA has released PersonaPlex-7B-v1, a full-duplex speech-to-speech model that replaces traditional voice assistant pipelines (ASR→LLM→TTS) with a single Transformer architecture. The model enables real-time natural conversations with precise persona control, supporting overlapping speech and natural interruptions.