The latest updates to OpenAI's ChatGPT include exciting voice dialogue options and the ability to communicate through images.

OpenAI Today, the announcement confirmed the deployment of advanced voice and image interaction features in ChatGPT, allowing users to chat vocally or explore images.

This revelation follows reports from Reddit users claiming access to OpenAI’s internal AI models, with one user sharing insights about a model dubbed Arrakis, which allegedly enables multi-modal inputs like text, audio, and video.

"The new voice feature utilizes an advanced text-to-speech model that creates lifelike audio from text combined with just a few seconds of voice samples,\" stated OpenAI. blog post We collaborated with professional voice actors to develop a variety of unique voice options. Additionally, we harness our own speech recognition technology, Whisper, to convert spoken language into text.

I came across some intriguing but unverified rumors regarding powerful internal models on Reddit.

– It's wise to approach these claims with skepticism. –

Apparently,
Two users have purportedly gained access to OpenAI's internal models and are sharing their findings on Reddit.

FeltSteam… pic.twitter.com/JRJH4xADZX

— Yam Peleg (@Yampeleg) September 25, 2023

Thanks to the newly introduced features, users can have dynamic conversations with ChatGPT using their voices, and they can also discuss and analyze images together with the chatbot. The rollout is set to occur over the next fortnight for Plus and Enterprise tier users.

Voice capabilities will soon be available on iOS and Android devices as optional features, while the image interaction is set to be launched across all platforms.

To engage the voice feature, users can navigate to Settings → New Features on the mobile app and select 'voice conversations.' Then, pressing the headphone icon on the top-right of the home screen lets users choose from a selection of five distinct voices.

For discussions involving images, users simply tap the camera button to take a photo or choose an existing one. On iOS or Android, they need to press the plus icon before proceeding. Moreover, users can initiate conversations about several images or utilize them to guide the chatbot's responses.

OpenAI states that image comprehension relies on multimodal models GPT-3.5 and GPT-4, which utilize their linguistic reasoning capabilities to scrutinize a wide variety of visual content, ranging from photographs to screenshots and documents that blend text with imagery.

OpenAI’s partnership with Spotify

Spotify also today announced this dynamic translation feature uses OpenAI's Whisper tool to transcribe English speech and convert other languages into English.

According to The Verge As part of its pilot initiative, the organization has partnered with well-known podcasters like Dax Shepard, Monica Padman, Lex Fridman, Bill Simmons, and Steven Bartlett to craft AI-driven translations in languages such as Spanish, French, and German for select episodes and upcoming titles.

"We are convinced that a considered approach to AI can forge stronger connections between creators and audiences, which aligns with Spotify’s goal of unlocking the creativity of humanity,” stated Ziad Sultan, Spotify's VP of Personalization.

Episodes translated into different languages by the pilot creators will be accessible globally to both Premium and Free Spotify users.

However, please keep in mind that the information on this page is not meant to serve as legal, investment, financial, or other professional advice. Always invest responsibly and seek independent financial counsel if uncertainty arises. For comprehensive details, consider reviewing the terms and conditions and support resources offered by the issuer or advertiser. MetaversePost strives for accuracy and impartial reporting, but market dynamics can change rapidly.