D-ID App Gives ChatGPT a Face and Voice

D-ID App Gives D-ID, the Israeli startup behind viral experiences like Deep Nostalgia, is giving ChatGPT a face and voice with its new web app. The generative AI chatbot will respond to queries typed or spoken through the app called chat.D-ID.

The app is still in beta with Alice as the single synthetic host, but D-ID plans to add other digital characters. It also wants to make it easier for elderly people to use AI.

Chat with a ChatGPT Avatar

Chatbots that respond to text queries have always been a popular way to interact with artificial intelligence, but now users can also talk face-to-face with photorealistic AI avatars. A synthetic media startup called D-ID has rolled out a new web app that lets people ask questions to an avatar powered by the company’s ChatGPT conversational engine.

The free-to-download web app, dubbed chat.D-ID, uses a combination of facial reenactment and advanced text-to-speech technology to give the avatars a more humanlike personality. To use the app, you simply type or click on a microphone icon and say your question to an avatar named Alice. Then, the avatar will answer your query with its own unique voice and personality.

D-ID’s facial reenactment uses a tool like Stable Diffusion to capture a real person’s facial expressions in video, which are then transferred onto the avatar using its own unique face animation technology. Then, it uses a text-to-speech model such as GPT-3 to translate your input into a natural-sounding speech pattern that the avatar can mimic.

Finally, the resulting photorealistic digital human is paired with an animated mouth and head movements by ElevenLabs to create an animated avatar that looks like it’s talking directly to you. Ultimately, the avatars can be used for both entertainment and practical purposes such as providing customer support or conducting a virtual job interview.

Talk to a Talking Avatar

D-ID is a cutting-edge generative AI platform that enables users to transform pictures and videos into extraordinary experiences. Its Creative Reality Studio and APIs provide a variety of features for use in marketing, events, e-learning, customer support, and more.

D-ID’s latest generative AI tool is a web app that allows users to talk face-to-face with a photorealistic artificially intelligent human. The web app, called chat.D-ID, uses real-time face animation and advanced text-to-video technology to create a photorealistic avatar that can respond to users’ questions. Users can ask questions by typing or saying their requests. The AI avatar then reads the text and produces a video response using voice synthesis and lip-syncing.

Avatar Pose stores the current pose of the avatar, capturing its emotions and physical state. Facial Expression generates a facial expression that matches the tone and emotion of the avatar’s voice output. Lip Sync converts the text into audio, ensuring that the avatar’s lips move in sync with its spoken words. The voice output is combine with the video rendering to create a high-quality, realistic video of an avatar speaking in any language.

This technology can be use for a wide range of applications, including enhancing video conferencing and improving personalization in e-learning/corporate training. It is also ideal for use in virtual assistants, providing customers with personalized content, and making presentations more engaging.

Chat with a Talking Avatar

D-ID has created a new tool that allows people to speak with an avatar using their own images and voice. It can be use for a variety of purposes, including enhancing video conferencing tools and creating personalized videos for social media.

The new app, which is available for mobile devices and desktops, uses D-ID’s chatGPT technology. Users can type their questions or click a microphone icon and say them out loud. The response will come from a photorealistic synthetic human called Alice. D-ID notes that Alice can answer nearly any question that’s posed to her.

In a press release, D-ID said it plans to use its generative AI tools for various applications, including assisting in virtual training and sales conversations and strengthening online support for domestic violence victims alongside Spring ACT. The company also wants to improve the quality of videoconferencing experiences with its avatars, which it believes can help people feel more at ease during meetings and conferences.

The new D-ID app is free to download for iOS and Android, but the company may charge once it gets more usage. The company will also launch a version for businesses with more customization options and higher video resolution. In addition, D-ID’s Creative Reality Studio enables businesses to add an AI presenter to any video from their own content.

Have Interactive Conversations with a ChatGPT Avatar

The app allows users to ask questions in voice or text and then see a virtual avatar with a face that matches the user’s photo. The avatar also speaks and explains the answers. It can be anything from helping with homework to explaining physics.

The machine generated the answers that uses deep neural networks to produce human-like text through transformer models. This is a subset of deep learning that produces text based on data it has seen before. such as transcripts and online text. Unlike other chatbots, these are generated on the fly and do not have to be pre-scripted.

D-ID says it is experimenting with new ways to interact with its generative AI technology. It is use in apps such as TikTok and Instagram. Its goal is to make it easier for people of all ages and abilities to use AI. That is why it has developed the chat.D-ID web app and the Alice avatar to give people a more human-like way to converse with AI.

While the new web app and a human-like avatar will help to reduce creepiness. it is still important for people to be aware of the limits of this type of artificial intelligence.


Bogost notes that ChatGPT answers often sound like student essays.

Leave a Reply

Your email address will not be published. Required fields are marked *