Create Podcast Visuals With Riverside AI Tools (Full Transcript)

[00:00:00] Speaker 1: I recently started a comedic conspiracy theory podcast with a friend and while we recorded the actual conversation ourselves, I still needed some visuals to help bring it to life and make it look more professional. Things like an intro, b-roll, and cover art. Normally I'd have to go and create all of those things across different tools and then bring everything back into my edit. But Riverside's new image and video generator lets you create and refine those assets without ever leaving your workflow. So I'm going to use it to create the visual package for our new podcast and walk you through the entire process. Let's get into it. Once you log into Riverside, select the image and video generator on the left hand side of the screen. From here you can generate an image or video by writing your own prompt or start with one of the suggested templates and edit it. You can even upload visual references to help the generator match what you're envisioning. You'll also see your previous generations if you have any so you can revisit and compare different versions as you work. Now the first thing I want to create is an intro for our new show, which is called the Blueberry Spaceship Podcast. So I'll click on the text box right here and I'm going to select video and I'm going to enter a prompt describing what I want. Create a cinematic intro featuring a blueberry shaped spaceship slowly drifting through space surrounded by stars and subtle blue and purple nebula clouds. Make it feel polished, playful, and mysterious with smooth camera movement. Include a write on title that says welcome aboard. Add mysterious background music and sound effects. Before generating the video, I can customize several settings here. First, I can choose from the available video models. Each model includes a short description explaining what it's used for, which makes it easier to choose the right one without already knowing the differences between them. The exact selection of models may change over time, but having several models in one place lets you choose the best option for the style, quality, or type of video you're trying to create. In this case, I'm going to do Sea Dance 2.0. I can also choose the duration, aspect ratio, and resolution. And since this is going through a widescreen podcast episode, I'm going to choose 16 by 9. I'm going to change my duration to 10 seconds and use the resolution that best matches my video quality, which in this case, 1080p. Some models also let you generate audio with the video. I'm going to leave that on since I want this clip to include sound effects or ambience. You can also turn that off and add your own sound effects inside the editor if that's what you'd rather do. I can also upload a reference depending on the model. An image can help establish the look of the scene, while a video can provide additional direction for the movement. For this first version, I'm going to go ahead and generate the video directly from this prompt. Now, the cost of the generation is shown here as well on the bottom right. Generating images and videos uses AI credits, and the amount depends on the model and setting that you choose. Riverside shows me the total before I click generate, so I know exactly how many credits it's going to use. Once I'm ready, I'm going to go ahead and click on generate, and that'll take a few minutes to generate depending on how complex the video is. Once my video is generated and looking how I want it, I can download the video or save it directly to the media folder of a Riverside project. So I got my video right here. Let's check it out. That's fun. Once this video is generated, you can just download the video and add it to whatever asset that you want, or you can save it to your Riverside media folder with this icon right up here. You can also import it directly into your video timeline by hovering over the video and then clicking on this filmstrip icon that says add to video. From here, you can choose which project or timeline you want to add it to. So then you just click on the project that you want, and it'll import into that timeline. Also, if the video isn't quite what you wanted, you can continue editing by clicking on the edit icon and then continuing to give it more prompts until it's shaped to exactly how you want. In this case, it's looking pretty good, so I'm going to go with this. Next, I need some B-roll. In one section of the podcast, we talk about aliens and why people are so interested in the possibility that we aren't alone. I want a visual that feels mysterious without going too far into science fiction. I'm going to switch to image and then enter this prompt, a mysterious light hovering above a quiet desert road at night, cinematic and realistic with a subtle blue glow in the sky. Just like with video, I can choose from several different models. Different models may be better suited for realistic images, illustrations, text, or detailed designs. I can also change the aspect ratio. In this case, I'm doing 16 by 9. And if I had an image with a composition or style I wanted, I could upload it as a reference. I don't need one for this, so I'll generate it directly from the prompt. I'll go ahead and hit generate. And again, this could take a few minutes depending on how complex it is. So here's the image that we have. This is looking pretty cool. That's exactly what I described. Once I'm happy with the images generated, I'll save it directly to our podcast project or import it directly into my project. And as always, you can continue to refine it, save it to your media, regenerate it, or download it. Now let's hop into the editor and put everything together. Now because I saved both assets directly to the project, the intro and B-roll are already waiting on the timeline, or I can just open my media folder and I can add the content on the Tidemime from there. From here I can trim the intro video, layer in music or sound design, add titles or graphics until it looks exactly how I envisioned it. Then I'll add the alien image over the part of the conversation where we discuss aliens. And here's how the episode looks with both generated assets in place. The conversation is still the focus. These visuals simply help introduce the personality of the show and illustrate what we're talking about. Now the final asset that I need for our podcast is cover art. And this time I'm going to use a reference image. I'll go back to the image and video generator and select image. Now I do want to mention there's already a template for show cover art here. So you can click on this and this will give you a template prompt. I'm going to upload some inspiration to give the cover art some visual direction of our podcast. This gives the model a reference for the colors, composition, and overall style. Now you can totally use the template prompt that we have in here, but for this example, I'm going to use this instead. Create square cover art for the Blueberry Spaceship podcast, featuring a blueberry inspired spaceship against a mysterious night sky with a fun conspiracy theory tone and space for a bold readable title. Because this is podcast artwork, I'll choose a square aspect ratio. So one to one and select the image model that I want to use. For this example, let's try chat TBT image two. It's going to cost 15 credits. I'm going to go ahead and hit generate. All right, so it's finally generated our podcast cover art. This looks freaking awesome. I'm not going to lie. I love the little additions here in the bottom left and question everything, especially gravity. It looks really cool. I love the spaceship shape that this looks awesome. And this was the first generation. I haven't done anything else. This looks great. Now, once I have the design where I want it, I can download it and make any final text or layout adjustments before uploading it through our podcast host. Now we have our new cover art right here in the podcast hosting section. Also one quick tip that I want to mention is that the more descriptive your prompt is, the better your result will usually be. If I know what I want, but I'm having trouble describing it, I'll go ahead and ask co-creator inside Riverside to help me write or improve the prompt before I generate anything. Or you can use something like chat TBT or Claude to help you get more descriptive. I'll also be including a list of prompt templates that you can copy and paste in the description down below of this video that usually gets me closer on the first attempt and saves me from doing as many iterations. Riverside's image and video generator gives you one place to create images and videos with access to all the big models, customize the settings, upload references, and refine your results. You can then download those assets or import them directly to a Riverside project and keep your edit flowing. For this new podcast, we used it to create an intro, custom B-roll, and cover art, all without leaving the rest of our workflow. So if you want to generate videos and images with Riverside, you can use the link in the description to get started. And if you want to learn more about recording, editing, and creating better content with Riverside, I'll link a few more videos for you to watch right up here and in the description down below. Be sure to like this video and subscribe for more content just like this. And if you have any questions, feel free to leave them in the comments and I'll be happy to answer them personally. Thank you so much for watching and we can't wait to see what you create.