OpenAI has introduced an Advanced Voice Mode for ChatGPT, which has garnered positive feedback for its impressive capabilities. Users have noted that the new voice mode features improved accents and can adapt closely to the user's voice, creating a more personalized experience. The technology reportedly excels in less common languages, outperforming previous text-based interactions. Additionally, the voice mode has been praised for its ability to sing, demonstrating the advancements in text-to-speech models and large language models over recent years. OpenAI is also experimenting with a feature codenamed 'hojicha', which focuses on response feedback and visual changes, alongside a new 'shimmer' feature for custom GPTs. Overall, the advancements reflect significant progress in AI voice technology, with users expressing excitement about its potential applications.
That's an ElevenLabs-level TTS, fully open-source, running on consumer devices! https://t.co/TvN02vSGBE https://t.co/4LH1cvSuY1
If you haven't tried OpenAI's Advanced Voice Mode yet you really should. It's crazy how human-like it can be at times. Absolutely wild that we flew past the turing test over the last couple of years and the world didn't really blink.
How is Advanced Voice Mode so much better than GPT text mode in less common languages? Completely different training set? It speaks Kimashami, Kimarangu, (of course Kiswahili) excellently... Where GPT struggles with the same phrases and interactions.