This article explains how your Asset Intelligence minutes are used when you apply Closed Captions, Auto-Tagging or Facial Recognition to video and audio assets, and how to avoid using minutes twice on the same file.
Table of Contents
How Minutes Are Counted
Video and audio files are counted in minutes of processing, based on the length of the file. A 10-minute video uses 10 minutes of your Asset Intelligence allocation when it is processed.
When a video is processed, IntelligenceBank analyses it once and generates everything in a single pass: the Closed Captions, the Auto-Tags (Audiovisual Keywords, Topics, Objects, Locations and Brands) and, if selected, Facial Recognition. This means:
- You are not charged per feature. Turning on Closed Captions, Auto-Tagging and Facial Recognition together for a 10-minute video uses 10 minutes, not 30.
- Closed Captions and Auto-Tagging share one allocation. Both count towards the same Asset Intelligence minutes.
- Audio files work the same way, using minutes based on their length. Facial Recognition does not apply to audio.
Images Compared to Video and Audio
Images and video/audio are measured differently:
| Asset type | Measured by | Using several features at once |
|---|---|---|
| Images | Number of images | AI Tagging, Facial Recognition and AI Asset Descriptions together count as one image. |
| Video and audio | Minutes of processing (the length of the file) | Closed Captions, Auto-Tagging and Facial Recognition together count once, for the length of the file. |
Turning On Features Later
Because Closed Captions and Auto-Tags are generated in the same pass, IntelligenceBank keeps the results even for the feature you didn't turn on. If you switch that feature on later, the stored results simply appear when you save. Facial Recognition is the exception: it needs the video to be analysed again.
| If the video has already been processed and you later turn on... | Extra minutes used? |
|---|---|
| Closed Captions | No. The captions were already created during the first pass and are shown once you turn them on. |
| Auto-Tagging (Keywords, Topics, Objects, Locations, Brands) | No. The tags were already created during the first pass and appear when you save. |
| Facial Recognition | Yes. The video is analysed again from the start, which uses minutes for the full length of the video a second time. |
Examples
| Scenario (10-minute video) | Minutes used |
|---|---|
| Uploaded with Closed Captions, Auto-Tagging and Facial Recognition all turned on | 10 |
| Uploaded with Closed Captions only, then Auto-Tagging turned on a week later | 10 |
| Uploaded with Auto-Tagging only, then Closed Captions turned on later | 10 |
| Already in the Asset Library, edited to turn on Facial Recognition and Closed Captions together | 10 |
| Uploaded with Closed Captions only, then Facial Recognition turned on later | 20 (10 + 10) |
Frequently Asked Questions
Does saving a video again use more minutes?
No. Editing details such as the title, description, folder or metadata does not reprocess the video.
What happens if I upload a new version of a video?
A new version is a new file, so it is processed (and counted) again if Asset Intelligence features are turned on for it.
Does uploading my own SRT file use minutes?
No. Uploading an existing SRT file adds captions without any AI processing.
How do I find out how many minutes I have, or add more?
Contact your Customer Success Manager to check your allocation, purchase more minutes or upgrade your package.
Need Help?
For more on each feature, see Video/Audio Auto-Tagging and Closed Captions for Audio and Video Assets. If you have questions about your usage, reach out to your Customer Success Manager or contact our support team.
Contact SupportWas this article helpful?
That’s Great!
Thank you for your feedback
Sorry! We couldn't be helpful
Thank you for your feedback
Feedback sent
We appreciate your effort and will try to fix the article