Google has made its new chatbot, Gemini Live, available for everyone to try out without needing a premium subscription. It can follow conversations better than before, even if you backtrack or correct yourself, which makes it feel more natural when you talk to it.
Right now, it only works in English, but other languages will be supported soon. You can also choose how its voice sounds, but it’s not a full replacement for Google Assistant yet, as it can’t help with your personal tasks.
Enjoy smooth almost human-like conversations with Gemini Live
When Google first launched Gemini Live, it was only available if you had a premium subscription to Gemini Advanced. After testing it, they found that Live is good with conversations, even if you change what you’re saying, make corrections, or do things people naturally do when talking.
Unlike older systems that might get confused in these situations, Gemini Live handles it smoothly and doesn’t lose track of what you’re saying. For example, if you’re asking Gemini Live to find information about a nearby park but decide you want to know about a coffee shop instead, you can cut the AI short immediately and say, “Actually, I meant Starbucks” Gemini Live will update its response to provide information without any confusion.
On September 12, 2024, Google announced that the AI is rolling out to all Android devices for free. They shared the news on X, accompanied with a 10-second teaser video. All that’s required of you to use Live is to set your phone or tablet to English (US) as the default language.
More languages will be supported soon. It should happen in the coming weeks, along with support for iOS devices. For now, it doesn’t integrate with any Google Workspace app. But it’s something we’d also like to see in the future, as well as a connection with Android Auto for a great driving experience.
Talking to your phone is more seamless than before
Gemini Live is Google’s version of ChatGPT’s Advanced Voice mode, although it’s not as smooth. While ChatGPT’s voice feature works directly with sound, Gemini Live first turns your voice into text, processes it, then turns the text back into speech. This method makes Live feel more like a quick fix rather than a fully polished solution compared to ChatGPT’s voice experience.
The AI is powered by the Gemini 1.5 Flash model launched in May earlier this year. It’s a new, fast, and efficient AI model that’s designed to handle high-volume tasks quickly and cost-effectively. Although it’s smaller than the 1.5 Pro model, it still processes text, images, and videos well. Plus, it excels in summarizing, chatting, captioning, and extracting data.
To use Gemini Live on Android, install the Gemini app from the Play Store. If you have already, update it to be on the latest version. Then open the app and tap the waveform with a twinkle icon in the bottom right corner to access the new feature. If you don’t see this icon, you don’t have it yet. Be patient as it’s still rolling out, and some users will experience delay in different regions.
