Overview

Some organizations do not have the luxury of being able to call out to the cloud at any time. For various reasons, organizations refuse to be exposed to the internet. To provide such organizations with AI insights within their infrastructure, VIDIZMO has utilized Azure cognitive services. Since cognitive services can be used with containers, VIDIZMO has the ability to use containerized cognitive services to utilize Microsoft's expertise in AI within the customer's on-premise infrastructure. VIDIZMO combines the Azure Cognitive Services with its all-encompassing, robust video content management application, to provide a powerful platform for enterprises to upload, store, secure, stream, search, share, and monetize their content as per their organizational needs.


Concept

VIDIZMO uses an organization's containers of cognitive services to generate AI insights from their digital content. This AI functionality will be done on the on-premise servers where containers of Azure cognitive services are deployed. Using advanced media AI capabilities and technologies, VIDIZMO indexer with Azure cognitive services provides enhanced video processing features that enable users to view, analyze and search hidden data within a video. This equips large enterprises with the necessary tools to fuel their content with the power of artificial intelligence. This is achieved by associating and indexing the video on the basis of the keywords, spoken words, faces, and other essential data extracted from within a video. 


Cognitive Services in Containers 

Containers are an increasingly popular technology with a big impact on business. Because containers abstract applications from their underlying operating systems and other services, these applications can run anywhere: in the cloud, on a laptop, in a kiosk, or an intelligent Internet-of-Things (IoT) edge device in the field. With containerized cognitive services, AI abilities can easily be part of your on-premise infrastructure without the need for AI engineers and ML researchers.


At VIDIZMO, we take containers a step further by including support for them in our modern video streaming platform. So, you can create a container of Azure Cognitive services and use it with VIDIZMO to generate AI insights from your digital media. You can take greater advantage and move technologies such as AI closer to the edge, where they can reduce latency and boost performance. 




VIDIZMO With Cognitive Services Provides You


Visual Text Recognition

Using OCR capabilities, text that is visually displayed in a video is captured and indexed to be made searchable in a video. OCR container needs to be deployed and accessible by VIDIZMO in your on-prem infrastructure.


Closed Captioning

Creates closed captioning files in VTT format using the closed caption containers on your on premise infrastructure. These CC files are downloadable and easily readable using our transcription pane. To know more, see: Understanding Transcription Pane


Multi-Language Detection

VIDIZMO can identify multiple languages in a single media file. All you need to do is to provide the language identification endpoint, i.e. language containers on your on-prem server. 


Multi-language detection works on the selected options of your chosen segmentation criteria. i.e. confidence-based or language detection-based. Confidence is when the transcript is segmented on the basis of high confidence for the detected language whereas language detection is whenever a language is detected in the speech.


If all of your media is in a single language, use single language identification. 


Note: For language detection, you need to deploy language containers and provide their endpoint is the VIDIZMO portal settings.


Audio Transcription

With VIDIZMO's transcription facility, you can generate transcripts by converting the speech into text by defining the languages you want to generate transcripts into. If your defined language is not found in your media file, the transcript will be generated in the language you have set as "fallback language".


Keywords Extraction 

Intelligent video processing techniques including semantic and logical analysis extract various keywords from the visuals and audio of the video to make it more searchable.


Configuration of Azure Cognitive Services in VIDIZMO 


i) Select the environment where you want the Cognitive services to run. i.e On-prem or On-Cloud

ii) Select the region where you have created the Cognitive services account, i.e. Global or Government (Gov cloud)

iii) Select the insights that you want Cognitive services to generate for your media

iv) Provide container endpoint. i.e. your on-prem server where the closed captions container is located (On-prem insights generation needs your container for closed captions)

v) Select the language of closed captions in your container. If it is not correct the closed captions will not be readable for VIDIZMO

vi) Select the language detection, i.e. Auto-detect or Auto-detect Multilanguage

vii) Segmentation criteria: Here you can select either "Based on confidence" or "Based on language detected". Confidence is when the trained AI model is confident that the detected language belongs to a specific phonetic category, whereas, language detected is when the language matches to one of your language containers.

viii) Language identification endpoint is the endpoint of your language container that is deployed on your on-prem server 

ix) In the detection languages field, you need to provide all the languages that are deployed on your servers 

x) If the language detection does not detect languages that you specified in the previous field, the insights will be generated in the Fallback language

xi) Provide the endpoint of the keywords container

xii) Provide the endpoint of the OCR container

xiii) Select the language for OCR

xiv) Select On/Off for automatic processing, if turned off no media will go under automatic AI processing

xv) Click on Save Changes to start using Azure Cognitive services with VIDIZMO