Today, Artificial Intelligence has revolutionized the way people have been looking at information collection and distribution. Easier uploading, targeted sharing and enhanced accessibility of content has had organizations realize how imperative it is to power their content with advanced media processing capabilities.
VIDIZMO offers you AI-powered visual and audio insights for smarter analysis and search optimization. To learn more about its offerings, see: Understanding Video Insights.
Here is how you may enable and view video insights in your VIDIZMO portal:
Before you start
Make sure you have video insights configured in your portal. To learn more about it, see: How to configure Video Insights in your portal.
How to Enable Video Insights
You need to enable the check for displaying video insights at the time of media upload. See How to Upload Media to navigate to the upload screen.
1. Click on the Advanced Settings tab located to the far right, which depending on the encoding provider chosen for the portal during setup, will have varying options for transcoding, transcribing and generating preview thumbnails.
i. Check Insights options from the list as shown below to enable advanced processing on the media.
Note: Make sure to enable this as soon as you upload a video, because you can not configure Advanced Settings for a media after it has successfully been uploaded and published on the portal.
2. Now, click on the Publish tab:
i. Check Display All Insights to be able to view those insights against your media on the playback screen. You may then proceed with other settings of your media.
How to View Video Insights
1. From the Portal's Homepage or Media Library:
i. Click on the media thumbnail against which you wish to view insights and you will be directed to Playback's Page.
2. From the Playback Page, click on the Insights icon beside closed captions on the pane being displayed adjacent to your video player. Here, you will be able to perform the following:
i. Search Insights for relevant labels, keywords, etc and navigate to that specific point in the video where they are being talked about.
ii. You can see a list of Faces that have been extracted from within a video. If many, you can expand the list to see more. Click on every face allows you to view various details regarding that person's appearance, such as his appearance ratio in the video against all other appearances in the form of a percentage. Moreover, selecting a face allows you to view all its occurrences mapped against time on a media timeline below. The points shown on the timeline allow users to navigate to that certain point where the occurrences have taken place.
iii. Next, you can view the Labels that have been associated with your video using AI. These labels correspond to generic aspects of your video, making it easier to identify objects/common scenes in your content. These labels being clickable, highlight which part of the video have they been identified in and helps you navigate to that point using the timeline.
iv. Here, you will be able to see a list of all Brands that were extracted from your content, either via visual text or speech processing. It checks your content against a predefined database of known and popular brands to associate them with your content for greater visibility and relevance.
v. In addition to that, you can view Topics inferred from the context of your content, and the flow of conversation/speech within the video. These topics are fairly significant when parsing and indexing the content based on what topics does it throw light on, or what is it that is being discussed here.
vi. Additionally, you can view the Keywords extracted from your video after a thorough analysis of its audio and video, trying to form logical connections and derive the gist or central idea of what is being delivered in the content. These keywords being clickable, highlight which part of the video have they been analyzed and extracted from and helps you navigate to that point using the timeline.
vii. Interestingly, it also gives you an effective, easy-to-comprehend Sentiment analysis report highlighting the emotional connotation of the content in question. It does that by taking the speech into consideration and carrying out a lexicon and semantic analysis of what is being said and done. To highlight the ratio of video that qualify under the three broad sentiments being Positive, Neutral and Negative, percentages are assigned to all three of them. These sentiments being clickable, highlight which part of the video have they been analyzed and processed from and helps you navigate to that point using the timeline.