Recording & Streaming

UNDER REVIEW

Introduction

This chapter covers the primary components, ICT architecture considerations, and best practices for planning, designing, and deploying a lecture capture system.

A lecture capture system consists of three key components:

  • Ingest is the capture of audiovisual sources and the creation of encoded digital video content

  • Management is the data storage, library categorisation, and video delivery of encoded digital video content

  • Playback is the decoding of the stored video data for viewing

Lecture capture systems may be integrated with standard AV equipment and many of the standards and best practices for commercial AV installations apply equally to a commercial lecture capture deployments.

Scope of this Document

This chapter focuses on lecture capture as it relates to the digital recording of video, audio, and related metadata or annotations. Lecture capture is primarily used for reviewing a lecture session after the fact, distance learning, or other education-related use-cases.

Other specialised recording systems such as multi-track audio recording, or CCTV security camera management systems are outside the scope of this document.

Glossary of Terms

Term

Definition

ABR;

Smooth Streaming

Connection speed, quality, and device types may impact the ability to stream video at certain quality settings, ABR automatically adjusts the video quality as the video plays to match the connection quality

Annotation

An image composition method for representing notes and markups of an image or video. Annotations are typically drawn freehand onto a touch interactive display, or using a stylus

Audiovisual (AV)

Encompasses the processing of audio and video components for presentation or recording systems. May include interactive components such as touch displays. May be automated by a control system

Bitrate;

Data rate

In computing, and transmission of data streams, bitrate is the number of data bits that are transferred per second

Codec

Hardware or software designed to encode or decode a digital data stream. Codecs may be used to encode a video signal so it may be saved as a video file, and codecs may be used to decode a video file or incoming data stream so that it may be viewed

CDN

Content Delivery Network (CDN) is a distributed network specifically designed for video streaming delivery for large audiences connected over the internet

Flipped Classroom

A blended learning strategy that reverses traditional learning environment where students watch recorded content and carry out discussions and research during the class session

FTP

File Transfer Protocol is a communications standard for sending digital files over a network

HDBaseT

A standard for the transmission of ultra-high-definition video & audio, ethernet, and control data over twisted pair cabling

HID

Human Interface Device is a peripheral that connects to a computer to allow a user to interact and provide input. Examples include keyboard and mouse

HLS

HTTP Live Streaming (HLS) is a web protocol for delivery of audio and video content to viewers via the internet

HTTP communication

Hypertext Transfer Protocol (HTTP) is the foundation of data communication for the world-wide web

ICT Architecture

Information and communications technology architecture encompasses the computers, servers, phone systems, and network switching and routing for an organisation’s network

IP Address

An Internet protocol address (IP Address) is used for network communication between computer devices. The IP address allows a device to be identified on the network

LAN

Local Area Network (LAN) is a network of interconnected computers within a building or department

Multi-stream Recording

Multi-stream recording (aka multi-camera recording) is a lecture recording method whereby multiple individual video sources are captured in real time and saved as a single recording session. During playback, all streams are frame-synchronised and viewers may freely switch between or overlay (PiP) the available video content streams

Picture-in-Picture (PIP);

Windowing Processor

A combined video image composed of two or more video sources. Combined images may be side by side taking up half the viewable area, or one image may take up the entire viewable area with the other displayed on an inset window positioned in the corner

QoS;

Traffic Shaping

Quality of Service is a method of prioritising specific, time-sensitive network data traffic to ensure it reaches its destination with minimal delay

Samplerate

In digital audio, samplerate is a measure of audio samples per second. High samplerates produce a high fidelity sound

SD Card

Secure Digital (SD) memory card digital storage media

WAN

Wide Area Network (WAN) is a computer network that spans over large geographical distances

Webinar

Interactive online seminar or lecture delivered to viewers via the internet

Stakeholders

Below is a list of stakeholders commonly involved in the planning, deployment, and maintenance of a lecture capture system:

  • AV and ICT Managers (design of integrated technology)

  • End Users (functionality, features, and user experience)

Define the Purpose

Common Objectives

Below is a list of common objectives for lecture capture:

  • Record presentation and class material of lectures

  • Record study information, discussions, and whiteboards from tutorials & studio sessions

  • Record detailed actions of instructor, or students (simulation lab) for forensic review

  • Peer-to-peer student-created video learning opportunities

  • Distance learning (remote students or remote instructor)

  • Flipped classroom (tutorials and discussions regarding recorded content)

Emerging Technology and Use-cases

A recording and streaming system may also be utilised for advanced applications:

  • Mobile recording (user’s device tablet / phone)

  • Live-streaming – Webinars (~10-50 viewers)

  • Live-streaming – Broadcast over the internet (100+ viewers)

  • Viewer participation through comments, annotation, et al

  • Speech-to-text video search and indexing

  • Artificial intelligence to generate close-captions (including language translation)

  • Artificial intelligence to computer speech translation (speech synthesis)

Audiovisual Room Types

Below is a list of common types of rooms where lecture capture may be deployed:

  • Lecture theatres

  • Auditoriums

  • Classrooms

  • Studios

  • Simulation labs (ie nurse training, flight simulators)

  • Activity-based learning spaces

Define the Scope

Ingest Method and Combined Sources

Define the video, audio, and data source requirements for input to the recording system:

AV Sources

Description

AV System Design Requirements

Fixed Ingest Sources

Video and audio source is permanent and consistent

Video and audio signals may be connected directly to lecture capture appliance (hardware) and/or computer with recording software

Switchable Ingest Sources

Video and audio sources may be any one selected from available sources. May have the ability to switch sources during recording session

Video: AV switcher, matrix switcher, Audio: Mixer, Audio Digital Signal Processor (DSP)

Fixed Combined Ingest Sources

Video sources may be combined (PiP) and audio sources may be mixed as video is encoded

Video: AV PiP or windowing processor

Audio: Mixer, Audio Digital Signal Processor (DSP)

Distributed Recording (Multi-Camera Recording)

Multiple video and audio sources may be recorded as a single session / stream. Viewers may switch between sources during playback

Video and audio signals may be connected directly to multi-input lecture capture appliance (hardware) and/or uploaded to video content management system (software)

Annotation – Video Overlay

Video content may be recorded with a digital annotation overlay

Video: Annotation video processor with HID input from touchscreen or keyboard/mouse

Content Management and Distribution

Define the method of managing and distributing recorded video content:

Content Management Method

Description

Video Content Management Requirements

User-Managed

Encoded video file and session data is saved to onboard storage, SD card, or USB storage. May be saved direct to lecture capture appliance or user’s computer

Distributed freely by USB storage media, email, shared drive, or cloud-based storage service

On-premises Video Content Management Server

Encoded video file and session data may be uploaded to a centralised on-premises video content management system

Web browser-based video content management and delivery platform.

Server software installed to on-premises computer hardware

Cloud-based Content Delivery Network (CDN)

Encoded video file and session data may be uploaded to a third-party cloud-based CDN platform (ie YouTube, Vimeo, et al)

Web browser-based video content management and delivery platform.

Server software accessed via cloud-based computer hardware

Live-Streaming – One-to-Many

Encoded video is streamed and recorded in real-time to a third-party CDN. The CDN manages distribution of the live-stream to audiences connected via the internet.

Live-streams may be saved as a video file for viewing after the fact

Web browser-based, video content management and delivery platform with support for live-streaming.

Server software accessed via cloud-based computer hardware

Webinar – Many-to-Many

Encoded video of instructor content is streamed and recorded in real-time to a third-party CDN. A group of connected users may view the instructor content, and interact in real-time. Connected users may interact with the group through chat, camera, voice, and/or file-sharing.

Webinars may be saved as a video file for viewing after the fact

Webinar content delivery platform facilitating instructor-led seminars over the internet and the capability for viewers to interact and engage with the group.

Server software accessed via cloud-based computer hardware

Data – Comment Sections

Users may comment on the video

Web browser-based video content management and delivery platform with support for comment sections

Data – Live Comments (Superchat)

Users may discuss or comment on a live-streamed video in real-time providing more context for their comments

Web browser-based, video content management and delivery platform with support for live-streaming and live-commenting

Viewing and Streaming

Define the method of viewing and streaming video content:

Viewing Method

Description

Viewing Requirements

Lecture Capture Appliance

Playback of encoded video file and session data through lecture capture appliance

AV system for presentation of lecture capture appliance output

Video Playback Application – Filesystem

Playback of encoded video file and session data video by opening the content in a video playback application installed on the end-user PC or mobile device

Video playback application installed on end-user PC or mobile device

Video Content Management and Delivery System – Web Browser-based

Playback of encoded video file and session data through web browser-based video delivery platform.

Content delivery server may be on-premises or cloud-based

PC or mobile device running a compatible web browser application (ie Chrome, Safari, Firefox, et al)

Recorded Content Creation Rate

Calculate the estimated creation rate of recorded content as follows:

    • Given N is number of rooms featuring lecture capture

    • Given H is amount of average hours of recorded content per day per room

    • Given R is the estimated total amount of recorded content per day

    • R = N x H

  • Example: If there were 5 rooms with lecture capture capability creating an average of 3 hours of video content per day, the organisation estimates that a total of 15 hours of video content will be added to its content library per day.

Data Retention Policy

Determine the length of time the recorded content may be available for viewing before it is archived or permanently deleted.

Quality Requirements

Define the quality requirements for encoded video recorded content:

  • Video codec

  • Image resolution

  • Frame rate

  • Colour bitrate

  • Video compression:

    • Data rate (Kbps, or Mbps)

  • Audio quality:

    • Option to remove audio track

    • Mono, stereo, or multi-channel

    • Uncompressed (WAV, PCM, AIFF)

    • Compressed (MP3, AAC)

    • Bitrate

    • Samplerate

The higher the quality of encoded video, the more storage space required per second of recorded content. An organisation’s decisions when determining the quality requirements for lecture capture should take into account any ICT architecture limitations (disk storage and data traffic transmission capacity).

Encoded Video File Size Calculations

Calculate encoded video file sizes by the following:

    • Given D is the encoded video data rate in Mbps

    • Given M is the encoded video length in minutes

    • Given F is the file size of video content in GB

    • F = D x M x 0.0075

  • Example: A 3-hour of video encoded at 7.35Mbps will be 9.92GB

The calculation may be inverted to conform to ICT architecture limitations.

Calculate total minutes of recorded content by the following:

    • M = F / (D x 0.0075)

  • Example: If the total disk space available is 2TB, and the data rate is 7.35Mbps then a total of 36,281 minutes (604 hours) of content may be stored.

Calculate optimum data rate by the following:

    • D = F / (M x 0.0075)

  • Example: If the total disk space available is 2TB, and 1,000 hours of content must be stored, then the optimum data rate is 4.44Mbps

Key Components

A lecture capture system is composed of three key components:

  • Ingest: Encoding and recording of video, audio, and data sources

  • Management: Content storage, transcoding, cataloging, and delivery

  • Playback: Decoding and playback of video data

Ingest Methods & Best Practice

Ingest covers the audiovisual sources, video encoding, and digital file creation aspect of a lecture capture system.

Video Encoding Hardware and Software

Lecture capture ingest may be achieved by hardware appliance, or software:

  • Hardware appliance common features:

    • Digital video inputs (HDMI, DVI, or SDI)

    • Analogue audio inputs (mic or line level)

    • Capture connected USB video and audio devices (webcam, microphones, et al)

    • Digital video output port

    • Line level output port

    • Ethernet port (video streaming over IP network, upload to shared file system or ftp, third party control)

    • Storage (onboard, SD card, or USB storage media)

    • PiP function (multiple digital video inputs)

  • Recording software common features:

    • Software running on computer device, or mobile device (tablet or phone)

    • Recording function may be included as a feature of Unified Communications or webinar software

    • Capture desktop (share screen)

    • Capture connected USB video and audio devices (webcam, microphones, et al)

    • Capture digital video input

    • PiP function

Multi-Stream Recording

Multi-stream recording (aka multi-camera recording) is a lecture recording method whereby multiple individual video sources are captured in real time and saved as a single recording session. During playback, all streams are frame-synchronised and viewers may freely switch between or overlay (PiP) the available video content streams.

Example: A recorded lecture session includes two available video streams. One is the presenter’s shared screen showing a slide deck, the other is a camera showing the presenter’s face. The slide deck may contain important visual information, by making a selection via the video playback software the viewer chooses to show only this and hides the video showing the presenter’s face.

This feature requires that the ingest hardware or software is compliant with the video content management system.

Encoded Video Best Practice

Video encoding best practice as it relates to lecture capture:

  • Confirm the video resolution, colour bitrate, and framerate matches the quality requirements for recorded content

  • Confirm the video encode data rate and compression matches the quality requirements for recorded content

  • Calculate the estimated hours of recorded content to be created each week. Confirm the total recorded content disk space requirements will not exceed the ICT architecture data storage or transmission rate as the library of recorded content grows

  • A copy of encoded video content may be archived at the highest possible quality to futureproof the video content library should the ICT architecture be upgraded

Video Capture Methods

Digital video capture best practice:

  • Confirm the camera, or video-outputting device specifications match the quality requirements for recorded content

  • Confirm any video processing or signal extension (ie HDBaseT) present in the signal chain will not compress, degrade, or downscale the recorded content

  • Confirm the video resolution, colour bitrate, and framerate matches the quality requirements for recorded content

  • High-bandwidth Digital Content Protection (HDCP) will block the digital video output (DisplayPort, DVI, HDMI) of a device if an unauthorised receiving device is detected. All lecture capture, and video recording devices are not authorised to record content with HDCP-enabled

  • It is illegal to use adaptors, convertors, or software to bypass or disable HDCP

Common sources and processing equipment deployed for digital video ingest:

  • Sources:

    • Cameras – pan/tilt/zoom

    • Cameras – fixed shot

    • Overhead document cameras

    • Webcams

    • Mobile device camera and microphone

    • Computer presentation content

    • Media playback devices

    • Video conference

  • Signal processing equipment:

    • Audio visual switcher / matrix switcher (switch between video source inputs)

    • Picture-in-Picture (PiP) / windowing processor (combines multiple video sources into a single video signal)

    • Annotation processor captures touchscreen or keyboard/mouse activity and renders a digital annotation overlay onto an incoming video signal

    • Video distribution amplifier (may be used to deploy a redundant video encoding device for critical applications)

Audio Capture Methods

Audio capture best practice:

  • Microphones:

    • When a single person’s voice must be heard clearly and distinctly, use lapel or handheld microphones with cardioid, supercardioid, or hypercardioid capsules. Position the microphone just below the person’s chin or clipped to the lapel.

    • When noise from the room must be captured whether it be a group discussion or room activity, use omnidirectional or multi-array microphones. These may be suspended from the ceiling or placed on a table centre to the general noise source area.

    • Where possible, reduce the loudness of any ambient room noise captured by microphones

    • Where possible, reduce acoustic reflections from sound within the room. ### Refer to Chapter on Acoustics ###

  • Line level Audio:

    • Ensure recorded audio is free of any hum or noise that may be caused by poor grounding

    • Where possible, use balanced line level for signal transmission

  • Use a noise-gate to filter out any ambient room sound

Common sources and processing equipment deployed for audio capture:

  • Sources:

    • Microphones (wired or wireless)

    • Audio playback devices (PC, media player, et al)

  • Signal processing equipment:

    • Audio Digital Signal Processor (DSP)

    • Audio mixer

    • Audio distribution amplifier

Video Content Management Systems

Video content management systems covers the file data storage, library categorisation, and video delivery aspect of a lecture capture system. Comprehensive content management systems may also include user access, transcoding, and metadata.

Video Content Management Best Practice

  • Categorisation

    • Define the video categories

    • Categorisation may be achieved by metadata features of the video content management platform, by specific naming of the video file, or by named folders on a file system

    • Categorisation may be by instructor, date, subject, class, or by student. The organisation is free to determine the best method of categorising the video library

    • A video library without categorisation will make it difficult for users to find the right content

  • Codec and file formats:

    • Use consistent file formats and codecs

    • If uploaded videos are of varying codecs, data rates, and quality settings consider transcoding each video file to a uniform format and quality

    • If the content management platform features Adaptive Bitrate (ABR), transcoding processes will be handled automatically

User-Managed Distribution Method

At the end of a lecture capture session, an encoded video file is created and saved to the connected storage of the encoding device. The file may then be distributed, edited, or deleted as required.

User-managed distribution considerations:

  • Files may be distributed by email, shared file system, or cloud-based file storage

  • Suited to small or infrequent lecture capture requirements

  • Access to video files, and intellectual property controls may be difficult to manage

  • File naming and quality standardisation and consistency may be difficult to manage

On-premises Video Content Management Server

Enterprise-class server software to manage a library of digital video content data.

On-premises content management server features and considerations:

  • Commonly accessed via web-browser interface

  • Accessed over LAN, WAN, or internet network connection

  • User accounts (software feature dependent):

    • Administrators may set video access controls per video per user

    • Viewing history

    • Preferences

    • Notifications

    • Comments

  • Facilitates categorisation and indexing of video content library

  • Includes search and browsing features to find indexed videos and content

  • Video file download to user’s device

  • Web-browser video playback features (software feature dependent)

  • Common video streaming protocols:

    • RTSP Real Time Streaming Protocol

    • RTMP Real Time Messaging Protocol

    • MMS Microsoft Media Server

    • HTTP Live Streaming (HLS)

    • MPEG-DASH (open standard used by Netflix, Google, Microsoft, Samsung, et al)

    • Smooth Streaming and ABR adaptive bitrate

    • WebRTC plugin-free peer-to-peer communication (not suitable for large, complex audiences)

  • Commonly uses SQL database for indexing and categorisation

  • Must be factored into ICT architecture design:

    • Video data rate and compression and how this may impact disk storage, and network traffic transmission rates

    • QoS and traffic shaping to prioritise video packets

Cloud-based Content Delivery Network (CDN)

Third-party video content management service and distributed network for video streaming. Facilitates video delivery via the internet and HTTP communication to large audiences.

Cloud-based CDN features and considerations:

  • Commonly accessed via web-browser interface

  • Accessed over internet network connection only

  • User accounts (software feature dependent):

    • Admin may set video access controls per video per user

    • Viewing history

    • Preferences

    • Notifications

    • Comments

  • Facilitates categorisation and indexing of video content library

  • Includes search and browsing features to find indexed videos and content

  • Video file download to user’s device

  • Web-browser video playback features

  • Live-streaming features (CDN platform dependent)

  • Webinar features (CDN platform dependent)

  • Common video streaming features:

    • Chunked delivery (segmented video delivery allows users to jump to any section of a video and begin streaming)

    • HTTP communication (allows information to be sent across internet, local LAN or corporate WAN

    • Nonlinear stateless interaction (client may request next video segment independently of the previous requested segment)

    • Cache-friendly (modern streaming works with HTTP caches)

    • Adaptive Bitrate (ABR) playback (connection speed, quality, and device types may impact the ability to stream video at certain quality settings, ABR automatically adjusts the video quality as the video plays)

  • Common Video Streaming Protocols:

    • HTTP Live Streaming (HLS)

    • MPEG-DASH (open standard used by Netflix, google, Microsoft, Samsung, et al)

    • Smooth Streaming and ABR adaptive bitrate

  • Must be factored into ICT architecture design:

    • QoS and Traffic Shaping to prioritise video packets

  • Reduces ICT architecture resource requirements

Playback Methods

Video content playback covers the file data delivery, decoding, and playback aspect of a lecture capture system.

Playback of recorded video content depends largely on the content management method, and ICT architecture and guidelines regarding the distribution of digital video assets. A user-managed distribution method may mean that video files are emailed to recipients to view on their personal devices. A video content management system may allow users to playback videos from a large library via a web-browser interface.

Hardware Appliance

A media player device capable of decoding and playback of digital video files and content. A lecture capture hardware appliance is commonly also capable of playback of recorded content.

Hardware appliance common features:

  • Digital video output port (HDMI, DVI, DisplayPort)

  • Line level output

  • Ethernet port (streaming, upload to shared network file system or ftp, third party control)

  • Storage (onboard hard drive, shared network file system, SD card, or USB storage media)

  • Limited video codec compatibility

Video Playback Software

The digital video files may be played directly from a file system and operating system of a computer device. Video playback software is used to open and decode the video file for playback on the user’s computer.

Video playback software considerations:

  • Video files may be stored on the user’s computer or USB storage media

  • Low impact to ICT architecture and network traffic

  • Playback of video files does not require network connection (once transferred to user’s computer)

  • Flexibility to use software for decoding and playback of any video codec

  • User access, and intellectual property controls may be difficult to enforce

Video Delivery via Content Management System

The Video Content Management System is accessed via web-browser interface. Playback of recorded video content is achieved in-browser by using standard HTTP communication. Recorded video content is accessed through a web-browser connecting over LAN, WAN, or internet.

Content Management System common features:

  • May be referred to as “Corporate YouTube” or “Private YouTube”.

  • Many content management solutions have adopted features that are commonplace for video sharing websites

  • Commonly accessed via user’s PC, or mobile device

  • Commonly accessed via web-browser application, although some content management solutions may feature an application for viewing video content

  • May integrate live-streaming capability via Content Delivery Network (CDN)

  • May feature playback of multi-stream recording sessions whereby the user may select from multiple video streams during playback

  • May include webinar instructor-student interactivity

  • May include Unified Communications features

  • User accounts

    • Admin may set video access controls per video per user

    • Viewing history

    • Preferences

    • Notifications

    • Comments

  • Consistent user experience for all users

  • Ubiquitous intellectual property controls may be deployed (ie video playback must be in-browser)