OpenAI Launches Advanced Voice Mode on ChatGPT Desktop App
OpenAI has officially rolled out its Advanced Voice Mode (AVM) to the ChatGPT desktop application. This update allows users to engage in real-time, hands-free voice conversations directly from their computers. Bringing advanced voice capabilities to the desktop enables users to multitask and interact with the AI assistant without needing a keyboard, moving closer to sci-fi-like virtual assistants. It enhances productivity by allowing seamless, low-latency voice control during daily desktop workflows. The feature leverages native multimodal speech-to-speech processing, which reduces latency and preserves emotional nuance compared to traditional text-to-speech pipelines. Users can access this feature directly within the macOS and Windows desktop applications.
## BACKGROUND
OpenAI's Advanced Voice Mode is powered by GPT-4o, a native multimodal model capable of processing audio, vision, and text in real time. Unlike older systems that transcribe voice to text, process it, and then synthesize text back to speech, GPT-4o handles audio natively. This architecture enables much faster response times and more natural conversational flows, including the ability to detect tone and respond with appropriate emotion.