ChatGPT has applied its voice conversation model, GPT-Live to make real-time dialogue sound more natural, as reported by Engadget on Aug. 9. The model is built to let it speak while listening to a user. ChatGPT's initial voice mode in 2023 worked through three systems: speech-to-text, a language model and text-to-speech, while the advanced voice mode later moved to a single multimodal model. OpenAI described GPT-Live as a "full-duplex" architecture, designed to reduce problems where earlier voice models replied before a sentence ended due to misjudging brief pauses. GPT-Live can intermittently offer short reactions such as "um" or "yeah" while the user is speaking, and can pass complex questions requiring reasoning or research in the background to models such as GPT-5.5. ChatGPT also has a new feature to create decorative text using Unicode character styles without installing separate font files, as reported by TechRadar. Users can enter a desired phrase and specify a style such as speech-bubble lettering, a gothic look or inverted letters. The text can be copied and pasted into other apps, but displays properly only in environments that support it; on some websites it may appear as empty squares or be replaced by other symbols. Highly decorative styles can reduce readability and may inconvenience accessibility tools such as screen readers, so it is appropriate to limit such text to titles or short emphasis phrases. ChatGPT noted that converting an entire long piece of