I used Google's Flow AI to create my own videos using sounds and dialogs.

You can generate videos from any desktop browser, but Google recommends using Chromium-based ones such as Chrome or Edge. Sign in to the Flow website with your Google account and click the “New Project” button. That step places you on a video generation page.

Click the settings icon to the right of the prompt. Here you can change the number of videos you want to view and the model you want to use. Usually, two default numbers are good choices.

Also: This interactive AI video generator feels like you're stepping into a video game – how to try it

For models, try one of the faster ones, especially if you're using an AI Pro plan and don't want to bite a lot of credits at once. Unless you plan on adding sound effects or conversations, you can also choose VEO 2 faster than VEO 2, as there are a fewer credits.

Assuming you're describing the video via text, set the option to text and set it to video. Next, enter a description of the video you want to generate. For help, I often ask ChatGpt to create a prompt for me. It provides basic information and AI adds details, but can be changed if necessary. Once you have added a description, click on the arrow to submit.

Also: With Hume's new EVI 3 model, you can customize AI Voices – how to try it

In my first video I submitted the following prompt:

“Vintage style video set in 1925. It shows men and women walking side by side along the sun-lit beach. The men wear straw boater hats, white button-down shirts with rolled up sleeves, and suspenders over highway trousers. One-piece dress with understated cuts, a striped 1925 swimsuit is a subtle, golden beach with gentle waves and soft lighting.

Flow creates a specified number of videos. When you're ready, use each video[再生]Click the button to watch. Hover over the video. In the floating toolbar, click the download button to download the video as its original size, upscale format, or animated GIF. (Yes, this is a cool way to create your animated GIFs.)

Also: This new AI video editor is an all-in-one production service for filmmakers – how to try it

Click the full screen button to view the full screen of the video. Click the 3-dot icon and[削除]to delete the video.

You can then generate a video using sound effects and dialogs. Click on the flow heading at the top to return to the main screen. Select a new project. To add sounds, choose one of the VEO 3 options, preferably VEO 3, faster, and avoid using too many credits. Add a description.

After that, you will need to specify the sound you want to hear. You can write things like “The music is playing in the background” or “I can hear the waves crashing along the beach.” For dialogue, identify which characters should be talking and identify “The man will quote women with words like, “Is it free for dinner tonight?”

Also: Best AI Image Generators of 2025: Gemini, ChatGpt, Midjourney, etc.

I have submitted the following explanation in chatgpt quick help about this:

“A scene from an epic film in a misty medieval valley at dusk. The knight in full silver armor rides a muscular carriage across rugged terrain. In the distance, a giant dragon with glittering red eyes and black scales lined up on a rock shelf. He swings his shining sword around with dragons.

Flow generates videos. Just like before, you can play each one, view either in full screen, download it, remove anything you don't want to keep.

You can also use still images to generate videos and create them on the fly or by uploading them. For this option, go back to the main screen, select a new project, select a model, then change the options above the text to video, frame, and video prompt.

Also: Bing adds Openai's Sora video generator – and it's free

Click the + icon at the bottom and then decide whether to generate an image or upload it from your computer.

If necessary, you can crop the image to fit the video format. Click to save. That image serves as a starting point for the video. Click the + icon again to add another image. Harvest and store again. That image will serve as the end point of your video. For this image, I added a turtle photo as a starting point and a cat photo as an ending point.

You can then select a specific camera angle and shot. Click the camera icon to select an angle or shot to include it.

Enter a description and send it. Here I told Flow to create a video of a turtle transforming into a cat. In response, Google generated a video that I would play and manage.

Finally, you can ask AI to create a video that incorporates many of these different elements, such as explanations, sound effects, and dialogue. That's what I did. In this video, we ask Flo to create a statue of a ballerina dancer who will come back to life in the museum's exhibition room.

I also test AI tools for my living. Here are the three image generators I use and how to do it

With the help of ChatGpt, I added some movie angles. This included a 360-degree panning, a face close-up when performing a pirouette in slow motion, with a low angle shot from the floor spinning round. For the sake of the sound, I told Flo to add emotional and orchestral background music that sounded like something from Tchaikovsky ballet. In addition to that, I asked the statue to come to life and tell the crowd: “Dancing brings life to me.”

Of the two videos generated, I was hit in the first video. Here, Flow got it right through all the elements and provided a short but beautiful, cinematic video.