How to combine pictures and words into story-telling videos using VideoPlus Studio

"A picture is worth a thousand words." Pictures and words can both tell stories, but combining them into a video can tell a more vivid story and bring static contents to life. We all like storybooks as kids. Words with pictures are more appealing and easy to recall. Now with generative AI, pictures can be generated from text, and text can be turned into speech spoken by an avatar. 

Learn how to transform an article into a dynamic story-telling video in this step-by-step tutorial.

1. Click 'Storybook' in the menu, then select the custom story generation method (last one) and paste the article, select 'GIF image and subtitle json' output format and submit the task. Download the resulting files when the task is completed.

2. Alternatively, use the contents from an article as prompts to generate some images using external AI image generators such as Stable Diffusion and Midjourney, pick images that fit the contents and organize them according to the narrative.

3. Click 'Presenter' in the menu, go through your presenters and decide which one to use to present the article. 

4. If you need to create a new presenter or update an existing one, select the right gender and language, listen to the audio samples of predefined voices to decide which one you like. To clone voice, select a voice good for cloning, use the 'Clone Voice' option then choose your audio file. The voice can also be fine-tuned by adjusting audio properties (pitch, speed and volume). 

5. If you also want an avatar to speak the text, select 'Use Avatar' for the presenter and define properties for the avatar. First the image, select a predefined one or use your own; then select the shape, location and height, for example, use shape circle to center-crop and surround the avatar in a circle, location 4 to occupy the top right corner, height 20 to occupy 20% of the video height; last decide if you want to use slightly moving head, that may look more natural for longer speeches, and whether or not to hide the avatar when not speaking.

6. Click 'TTS for Document/Image' in the menu. If you have the Storybook output, open the GIF file and import the subtitle file using the function buttons in the top toolbar. Otherwise open a generated image. The first image also serves as the guiding frame that other images will have to fit in, resize if needed. A subtitle is associated with every image, the text will be spoken by a presenter and the image will show for the duration of the speech. 

7. If no Storybook output is used to populate all the images and subtitles, you'll need to add more images and fill in the subtitles with text corresponding to every image, also select the presenter.

8. You can add or remove an image, move the images up and down, and replace an image with a different one. You can always undo a change using the 'Undo' button in the top toolbar if you changed your mind.

9. Go over the whole article until you're happy with the contents showing in both the left and right panels, then use the export function button in the top toolbar to save the subtitles to a local file in json format, so that it's easy to make adjustments later if needed. Next to the "save subtitles" button there's a "save images" button that can pack all the images into a GIF image and save it to a local file, this is convenient for reloading all images to make adjustments.

10. Next to the "save subtitles" button there's a "save images" button that can pack all the images into a GIF image and save it to a local file, this is convenient for reloading all images to make adjustments when needed.

11. Click 'Submit Task' button in the top toolbar, review the options and set the values as appropriate. If the video is not short, you may want to specify a short Output Length first, for example, 10, to generate a 10 seconds video as a preview, before generating the whole length video.

12. Submit and wait till the task is done (notified by email), or check its status in the task history, check back if not ready.

13. Find the task and download the resulting video when available, see if you should adjust some text, presenter, etc., if yes, submit the task again after making changes, repeat the process until getting the desired result.

Following is the sample resulting video: