Vibe coded with Gemini 3: With, a video recorder that talks back
By Zara Zhang
Summary
## Key takeaways - **Monologue Becomes Two-Way Conversation**: It's a video recording tool where the AI actually talks back at you, so your monologue becomes a two-way conversation. [00:22] - **Gemini Made Development Easy**: Honestly, not that much. Like, Gemini is so powerful, it mostly just worked. [00:34] - **Use Cases for Content and Practice**: This is very useful for video content creators, for social media, also for like preparing for job interviews, and then practicing speaking when you're learning a language. [00:45] - **One-Click Silence Removal**: I can actually click remove silence with one click. It's going to remove all the silent parts in the video where I wasn't speaking, for example, the parts when the AI was speaking and I was silent. [01:07], [01:13] - **Video Edits from 1:15 to 0:31**: So now this original video is 1 minute and 15 seconds. Now after the silence is removed. It's now 31 seconds. [01:23], [01:30]
Topics Covered
- AI Transforms Monologues into Conversations
- Gemini Edits Videos Effortlessly
Full Transcript
Introducing Width. It's an application that helps you record videos with live prompting from the AI to get your creative juices flowing. First, I can choose the dimension of the video. Then,
I can choose voice mode or text mode.
Let's choose voice.
I just built a really cool application with Gemini through Vibe Coding.
>> That's fantastic. What kind of application was it?
>> It's a video recording tool where the AI actually talks back at you, so your monologue becomes a two-way conversation. Wow, that sounds really
conversation. Wow, that sounds really cutting edge. What was the biggest
cutting edge. What was the biggest challenge you faced while developing it?
>> Honestly, not that much. Like, Gemini is so powerful, it mostly just worked.
>> What kind of situations do you see this application being particularly useful for? Who would benefit the most?
for? Who would benefit the most?
>> This is very useful for video content creators, for social media, also for like preparing for job interviews, and then practicing speaking when you're learning a language, etc. Those are some
really compelling use cases. Once I stop recording, I'm going to get the video file that I just recorded, which I can download as an MP4 file. But here's
where it gets interesting. I can
actually click remove silence with one click. It's going to remove all the
click. It's going to remove all the silent parts in the video where I wasn't speaking. For example, the parts when
speaking. For example, the parts when the AI was speaking and I was silent. I
didn't want those parts in the video.
And I was shocked that I could implement this feature in Gemini and it just worked. It could actually edit the
worked. It could actually edit the video. So now this original video is 1
video. So now this original video is 1 minute and 15 seconds. Now after the silence is removed. It's now 31 seconds.
Let's play it. I just built a really cool application with Gemini through vibe coding. It's a video recording tool
vibe coding. It's a video recording tool where the is now a very clean video with all the silence cut out and I can download the new video as a file. Let's
try out a different dimension with the text mode. I'm participating in the
text mode. I'm participating in the Gemini VI coding hackathon.
So the prompt is going to display as live text on screen without
Loading video analysis...