Today, companies have large teams of trained workforces to perform tasks such as finding where the end credits begin in an episode, finding ad spots, or breaking up videos into smaller clips for better indexing. These manual processes are expensive, slow, and cannot scale to keep up with the volume of content produced, licensed, and even retrieved from archives daily.
VOD (Video on Demand) is everywhere. More and more people want to watch their online videos whenever and wherever. This is why accessing content online from video libraries is the new norm. Traditionally broadcast channels and media outlets forced consumers to view content on their schedule. VOD services allow audiences to access videos at their own time and from any compatible device. The meaning of VOD streaming is different from live video streaming because it is previously recorded as opposed to broadcast in real-time.
Netflix, Hulu, Disney+, and HBO are some examples of streaming services that use the VOD model. Many VOD platforms use OTT streaming, which means they stream over the internet. To start leveraging VOD streaming for your business, you’ll need to know the QC requirements for each streaming service.
What to Watch Out For Before Delivering to VOD Platforms
Before you deliver your project to any VOD platform, make sure you have checked the following areas for QC. Commonly QC checks baseband quality, compression artefacts, file integrity, and standards compliance. Content must be reviewed according to multiple quality parameters for baseband quality checks before final delivery. A comprehensive QC tool checks baseband quality, such as video signal levels, blur, defective pixels, black frames, colour bars, RGB colour gamut, mosquito noise, audio levels, audio clipping, and loudness levels.
When the content is compressed, some artefacts like blockiness, pixelation and ringing may occur in your content. Broadcasters will look for a QC tool that keeps transcoded content artefact-free. File integrity and compliance checks ensure that the delivered file or content is not corrupt. If these requirements are not met, downstream tools may not play the assets accurately.
The Different Types of QC for VOD Services
Each distribution network has distinctive QC requirements that you need to know. These distribution networks have documentation that outlines all of their requirements before publishing your content. Here are the top 5 most extensive networks and their specifications for QC.
Netflix outlines three different types of QC: QC Operations, Branded Content QC and Localisation QC. Their document explains common asset failures and provides further context to any redelivery requests their partners may have when delivering content through their QC processes. You can find the documentation here.
QC Operations: Identifies issues that would be disruptive to the streaming experience. In other words, they make sure the customer experience is simulative. This includes issues that would prevent consumption (e.g. incorrect content), issues that reduce the quality of the title (translation errors, typos, video artefacts) and maintaining consideration for licensed content. The delivery process for content going through QC Operations consists of three steps: Asset Upload, Auto QC/Inspection as a Service (IaaS), and Manual QC.
Branded Content QC: Validates Source Asset quality against Netflix Originals Delivery Specifications and checks for content consistency. Branded Content QC operators perform multiple total linear reviews of assets and flag issues that are inconsistent in the context of the program, such as technical issues (frame issues, aspect ratios) and content issues (audio ticks, double pixels).
Localisation QC: Qualifies translation quality, consistency and style guide conformance. This process involves a QC operator reviewing the timed text asset, implementing changes and categorising the reasoning for those changes, such as subtitles and closed captions.
Amazon has quite an extensive list of delivery specifications. To help with this, Amazon has its own machine learning powered video analysis service - Amazon Rekognition Video. This API makes it easy for media customers to automatically detect frame-accurate end credits, black frame segments, shot changes, and colour bars in video files using machine learning. Here is what you need to know about quality control for VOD platforms using Amazon Rekognition Video:
Black frames detection: A short duration of empty black frames with no audio are usually used as spaces to insert advertisements or to separate the end of a program segment, scene or the opening credits.
Opening and end credit detection: Automatically identifies the exact frames where your project's closing credits start and end. With this information, you can generate markers for interactive viewer prompts such as ‘Next Episode’ in VOD applications or find out the exact last frame of your footage. You can also generate ‘binge markers’, or interactive viewer prompts such as ‘Next Episode’ or ‘Skip Intro’ in VOD applications.
Shot detection: A shot is a series of interrelated consecutive pictures taken by a single camera for continuous action in time and space. You can detect the start, end, and duration of each shot and count all the shots in a piece of content. This helps you to create promotional videos from selected shots, generate preview thumbnails and insert ads in spots that don’t disrupt the viewer experience, such as the middle of a shot when someone is speaking.
Colour bars detection: By detecting sections of video displaying SMPTE colour bars, you can detect specific patterns to ensure colour is calibrated correctly on different screens from broadcast monitors, programs, and cameras.
Slates: Slates are sections, typically at the beginning of a video, containing text metadata about the episode, studio, video format, audio channels, and more. This helps identify the start and end of such slates, making it easy for operators to use the text metadata with further analysis.
Studio Logos: Studio logos are sequences that show the logos of the production studio involved in the production. You can identify such sequences and review them.
Content: Content refers to the portions of the TV show or movie that contain the program or related elements. This enables operators to internationalise the content, and customers can apply additional domain-specific rules.
Disney Streaming Services (DSS) has Transmission Operations Centres (TOCs) that monitor the quality control on any of the thousands of IP-based streams that they might be processing at a given time. DSS have outlined their media technical specifications into three sections: production, mastering and localisation. Find details for each specification here.
Production: These describe the technical requirements for the content to successfully be delivered to the mastering facility for the final QC pass. These are primarily for production facilities that are creatively finishing content. The following segments are:
Video: various image-focused specifications
Audio: different audio-focused requirements
Naming Convention: a method for naming media assets
Preliminary Materials: early production deliverables
Archival Elements: synopsis of Disney+ originals' archival media requirements
Acquisitions: titles' requirements
Mastering: This section describes both the lossless masters and the distribution masters that a mastering facility may be asked to create. These are the elements made from the final production materials.
Video: various image-focused specifications
Audio: audio requirements for delivery into the Disney ecosystem
Naming Convention: naming media assets
Timed Text: subtitle specifications, style, and related guidelines
Appendices: guidelines outside of typical deliverables
Resources: task-based files for download
Theatrical Mastering: temporary home of Theatrical Mastering specifications
Localisation: Elements that will be for localisation purposes.
Video: localised video elements
Audio: deliverables for audio dubbing
Viacom outlines Media Delivery Technical Specifications for every VMN US Network Operations, and their technical standards and practices apply to your content delivery. They have separated technical requirements for both SD and HD content. You can find more detailed information here.
Video Standard: HD video content must be delivered as 1920x1080 16x9, 10-bit, 4:2:2 component digital video at 29.97 frames per second.
Video Levels: Specifications for luminance, chrominance and black levels.
Audio Standard: Audio content must be delivered as 16-bit, 20-bit, or 24-bit uncompressed (PCM) digital audio at a 48khz sample rate.
Audio Levels: Outlines dialogue, peak audio levels and other objectionable artefacts.
Audio Channel Assignments: How audio signals appear on Stereo Left and Right.
Closed Captioning: How to provide closed caption data.
To help facilitate the delivery and management of content, Hulu has created the Content Portal. More detailed specifications from metadata to final subtitles can be found here.
Content Upload: The Content Portal can upload art for series, assets, film and channel art.
Asset Management: This allows you to view your deliverables, receive feedback, check your status and make any changes to your content.
Traffic Metrics: You can filter and export your data by monitoring your content metrics (views, time watched, view throughs).
Finance reports: If you are paid by an advertising revenue share, you can see financial reports for your content on a per-asset basis.
There are many different video distribution networks to choose from, and consumers are streaming more content tha77n ever. Expectations are high, and there is no excuse to deliver lesser quality on any device. Delivering QC on every screen is crucial for maintaining subscriber satisfaction, driving customer engagement, and increasing profitability.