In one of the most exciting ChatGPT updates since it became possible to install ChatGPT plugins, the popular AI chatbot now lets you prompt it using voice commands and by uploading pictures.
It means that ChatGPT is now much more than just a text-based companion for all your online endeavors, with the addition of these interactive features moving it one step closer to being able to replicate human-to-human interactions.
The new functionality will initially be available to ChatGPT Plus and ChatGPT Enterprise users, who will get access “in the next two weeks.” This suggests a phased rollout, and if you want to check if you’re one of the lucky ones who can talk to the generative AI tool early, here’s how to use voice commands with ChatGPT.
How To Use Voice Commands on ChatGPT
OpenAI, who owns ChatGPT, announced the blockbuster update in a recent blog post. The voice update will be available only on the ChatGPT apps for Android and iOS, while image-based interactions will be supported on the web platform, too.
If you’re a ChatGPT Plus or Enterprise subscriber and want to check out the new feature for yourself, then it couldn’t be easier to do so. Simply open ChatGPT and make sure you’re logged in to your premium account, then navigate to Settings > New Features and you’ll be able to opt-in to voice conversations. Remember, if you can’t find this option right now, it’s probably just because it hasn’t rolled out to just yet and there’s no need to panic.
Once you have secured access, it’s all about learning to how to use voice with ChatGPT to the best effect. Bearing in mind it’s mobile-only, you’re likely to get the most mileage simply from the convenience of being able to ask it to execute the best ChatGPT prompts, instead of having to type them. It’ll be fascinating to see how people do interact with when it’s more widely available, which is something OpenAI says will happen in due course.
This just in! View
the top business tech deals for 2024 👨💻
How To Use Image Input on ChatGPT
The other major addition to ChatGPT is the ability to input images into the chatbot. The use cases here are more obvious, given the visual nature of the society we live in and OpenAI has already offered up a handful of suggestions on how to use the new feature.
“Snap a picture of a landmark while traveling and have a live conversation about what’s interesting about it. When you’re home, snap pictures of your fridge and pantry to figure out what’s for dinner (and ask follow up questions for a step by step recipe). After dinner, help your child with a math problem by taking a photo, circling the problem set, and having it share hints with both of you,” the company writes.
There’s no need to opt in to this feature, apparently, it should just become available to Plus and Enterprise subscribers over the next couple of weeks.
How Does Voice on ChatGPT Actually Work?
For those interested in the more technical aspects of how voice works with ChatGPT, what we know right now is that the capability is powered by an all-new OpenAI text-to-speech model that’s “capable of generating human-like audio from just text and a few seconds of sample speech.”
OpenAI adds that it’s working with professional voices actors to create the different voice profiles that will be offered when using the feature, while transcription of a user’s words into text is made possible by the company’s Whisper system.
After rolling out to Plus and Enterprise users, OpenAI says that developers will be the next to get access to ChatGPT’s new voice and image input powers.
