In many science-fiction movies, the presence of virtual AI assistants is often used to create a futuristic atmosphere. Iron Man had Jarvis, Her had Samantha, and there are many other movies that have virtual assistants. And when ChatGPT launched in 2022, people felt that the age of AI had arrived. The AI ​​chatbot could have human-like conversations, write poetry and prose, compose music, and more. But it couldn’t actually “talk” to people because there was no voice mode for ChatGPT. This feature, still in the works, has been delayed and is expected to launch later this month. But what if there was an AI chatbot that could actually talk to you in real-time and answer all your questions? Moshi, an innovative chatbot from French AI company Kyutai, has arrived.

Moshi said hello

Unlike its competitors, Moshi is designed to engage in natural conversations while interpreting emotional nuances in your voice. This unique capability sets it apart, promising a more personalized and empathetic user experience. While OpenAI is working on refining ChatGPT’s voice features, Moshi is already ahead, offering a glimpse into the future of interactive AI technology.

The AI ​​voice assistant can handle two audio streams simultaneously, allowing it to listen and respond at the same time. Moshi’s launch was recently live streamed and has been making headlines ever since. And this new AI voice assistant is available to try for free.

How to use Moshi

To talk to Moshi you just have to go to us.moshi.chat. After opening the website you will see a black screen on which a message will be displayed.

In this message it is written, “Moshi is an experimental conversational AI. Don’t take everything it says seriously. Conversations are limited to 5 minutes. Moshi thinks and speaks at the same time. Moshi can listen and talk all the time: maximum throughput between you and Moshi. Ask it to play some pirates, how to make lasagna, or what movie you last watched. We try to support all browsers, Chrome works best. <3 Baked with @Kyutai. You're on the US demo. Depending on your location, maybe the EU demo will provide better latency."

Just below this message there will be a box asking you to enter your email. Enter any email ID and click on “Join Queue”. After doing this, another screen will open in front of you. On the left side there will be a speaker which will light up when you say something and on the right side there will be a box in which whatever Moshi is saying will be displayed.

For now, you can talk to the AI ​​voice assistant for a maximum of 5 minutes. At the end of the conversation, you can download the video or audio of your chat with Moshi by clicking on the option that appears at the bottom left.

If you want to start a new conversation on another topic, you can click “Start Over” at the top of the screen.