OpenAI has launched Voice Command and Image Upload features for ChatGPT Plus

OpenAI has unveiled new features for its AI chatbot, ChatGPT, including voice commands and image uploads. Initially, these features will be accessible to premium users (ChatGPT Plus), with a wider release scheduled later.

The voice command feature enables users to ask questions verbally, which are then transcribed into text for the chatbot to process and respond to in spoken form, largely using Whisper technology for this. ChatGPT will offer users a selection of five synthetic voices, and while the company recognizes the potential of synthetic voice technology, it also acknowledges risks like impersonation or fraud. Therefore, it plans to restrict and regulate the use of this technology to certain cases and partnerships, such as their recent collaboration with Spotify to use its text-to-speech technology for translating podcasts into multiple languages, while preserving the original podcaster's voice.

In the other hand, the chatbot's new image search feature lets users upload a photo and receive responses based on the image, kinda like Google Lens goes. Users can also use drawing tools or text and voice queries to specify their image-based questions, enabling a dialogue with the AI. OpenAI has deliberately curtailed ChatGPT's capacity to analyze and comment directly about people for accuracy and privacy reasons.

As mentioned at the beginning, the new update will be rolling out over the next two weeks, during which subscribers to the Plus service will be able to engage in voice conversations with ChatGPT (iOS & Android) and include images in their conversations (all platforms).