Opis
Train your own AI system with VSN AI
With VSN AI you can have and train your own Artificial Intelligence engine to achieve much more specific and reliable results in image and audio analysis. The ability to train the system in a customized way to automatically detect a specific type of content with a high level of accuracy multiplies the applications and uses of AI in Broadcast and Media. For example, this feature is especially relevant for content moderation, broadcast monitoring, automatic translation of certain languages or for the automatic detection of celebrities and public figures relevant to a specific country, region or municipality.
Wide range of information
From the moment that a media file is ingested, users can automatically extract a great amount of data and information from the advanced image processing features (facial recognition, objects,
places, organizations and logos), audio layers analysis (music, voice, speakers and audio effects), speech-to-text and automatic translation, Optical Character Recognition (OCR) or sentiment analysis or emotions’ detection, among many others.
places, organizations and logos), audio layers analysis (music, voice, speakers and audio effects), speech-to-text and automatic translation, Optical Character Recognition (OCR) or sentiment analysis or emotions’ detection, among many others.
- Automatic content cataloguing
Users can focus on content creation and news production, leaving the immediate cataloging of media to the its AI tools. They would only need to perform a quick control and validation of the metadata obtained, saving time and resources that can be reallocated. In addition, with AI they can obtain a higher volume of information, which is also more accurate and leads to a more precise and exhaustive categorization of content.
- Advanced search functionalities
The volume of data generated from each media asset allows users to perform more accurate searches thanks to the big amount of options and parameters available. This way it is easier for users to adapt the search of content to their specific needs, reducing the number of results delivered by the system, which will only focus on those truly relevant to the user according to the keywords provided.
- Quicker content creation
Thanks to the integration of Artificial Intelligence engines within VSNExplorer, it is now possible to automatically detect and retrieve specific video segments that comply with a set of given parameters and metadata, drag and drop these segments directly over VSNExplorer’s web video editor, Wedit, and consolidate a final and already edited piece ready for broadcast, all without changing between windows.
- Translation, captioning & subtitling
Subtitling process can now be automated thanks to language analysis, speech-to-text and automatic
translation functionalities of AI-tools. These capabilities allow for the automatic translation and generation of captions in several languages of audio texts, delivering the corresponding text distributed by segment with mark-in and mark-out time codes.
translation functionalities of AI-tools. These capabilities allow for the automatic translation and generation of captions in several languages of audio texts, delivering the corresponding text distributed by segment with mark-in and mark-out time codes.
- Content moderation capabilities
Content moderation is sometimes mandatory depending on the legislation applied in the region where some specific content is being broadcasted. Sentiment recognition systems, image and objects analysis, and natural language analysis enable the automatic detection of sensitive or adult content, both within image and audio
layers, so that they may be adapted or eliminated directly using a video editor like VSNExplorer’s Wedit.
layers, so that they may be adapted or eliminated directly using a video editor like VSNExplorer’s Wedit.
