| |
|
An online presentation using the new Rich Media approach draws on exiting technology and web methods already well established in modern workplace,
lecture presentation and e-learning environments.
Professionals with a need to present to live audiences use
PowerPoint, video and/or audio, section title headings and various online delivery methods
such as web streaming as a normal part of everyday life.
Using Rich Media technology, it is now possible to incorporate
into browser-based online delivery output, fully synchronised video,
PowerPoint sets, PDFs, pre-bookmarked web URLs, electronic whiteboard
output, fully working Flash animations, Flash Paper documents, other
image formats and so forth.
The technology to harness the full effect of this has made a huge leap in the last year, with Accordent's
new Capture Station being internationally recognised as the state of the
art approach.
Accordent use proprietary code within the presentation software to
link the different elements together so a viewer can have as complete
an experience of the presenter's lecture as possible.
The purpose built multi-input encoder 'Capture Station' makes this
possible by generating the output result in standard web code, viewable
directly - and behind firewalls - in Microsoft Internet Explorer as well
as other browsers set up to play Windows Media Video and Real Video. The
result is also compatible with the Blackboard online delivery system. A
special feature for e-learning specialists is the ability to automate
upload of the presentation to Blackboard.
We feel it is important to distinguish here between the 'ease of use'
for SMEs (subject matter experts) of a product like Capture Station and
'ease of initial technical set-up'. We are convinced that in general,
installation and set-up should be completed by those trained to do this
-
prior to an SMEs' introduction to its use.
There are many skills required to set up an efficient Rich Media lecture
environment. Automated cameras, lights, sound equipment, streaming
server URLs, a fully functioning web portal... This collection of skills
is usually not what SMEs are paid to be experts in. However once initial
set up is complete, operation can essentially be run as a
one-button-begins-all process. Alternatively the software can be set to
run at prescheduled start and stop times, meaning no presenter
input required. Or run from a centrally located room - suitable for
where several Capture Stations can be scheduled live or automatically
from a single Accordent Media Centre control module - where one operator
controls all.
Whatever is fed into the lecture room projector (laptop - including
mouse movements, electronic whiteboard etc) is automatically captured in
high resolution and synchronised to the video and audio components.
This data is then simultaneous auto-uploaded to a streaming server (for
video) and web server (for content). The URL supplied to viewers automatically
links these two components together again, for seamless viewing in a
normal browser window.
Most components can be edited later - for example a PowerPoint slide may
need to be changed and so forth. Pre or post presentation, webcast
schedules can be auto-circulated to a secure selected viewer list.
This list can then be monitored to see who watched, what they watched
and for how long. A test if required, can be auto-offered and marked as
part of Accordent's Media Management software. 'Email The Presenter' and
'Download Additional Material' sub-windows are fully functioning during
the presentation. All within the one software package!
When I began producing Rich Media in 2000, it was do-it-yourself coding
linking video, text and index using the SMIL language. Every element change required an HTML text editor. It is not that difficult in an empirical sort of way, however mind-numbingly time consuming and not suited to the
academic or executive-on-wheels class at all.
Trying out Accordent's automated approach to doing all this made such a difference to
my way of thinking. Wizard and template based setup with detailed control panels. The ability to get going fast - around 1 minute from boot up to encoding a new
multi-input presentation for example, means a lot when many sessions are scheduled in a day.
If the power fails - we did a job using Presenter Plus in Italy not that
long ago where the power went off many times a day - the entire
presentation remains intact. Using Windows Media Encoder, the video is
OK, bookmarks and assets still in place etc. Very reassuring. The
software is bullet-proof.
Each Accordent product comes with many pre-built template 'skins' offering a range of features and looks. These can be easily altered in the software itself
for logo and other changes. For a more radical make over, a graphics program such as Photoshop and an HTML text editor
can give unique colour schemes and so forth if the myriad of different
template looks provided just aren't quite enough...
In any Accordent presentation, almost all the files are simply auto-copied from the chosen skin template folder, into the new encoding project folder either in an 'upload-to-web-server' process, a 'save-to-local-folder' process or both.
There is no rendering time involved, as the output appearance is all
HTML. The video
and/or audio stream is encoded either direct to a dedicated streaming server routed by the software, to the above local folder, or to both if required.
The additional 'assets' component, either in the form of pre-captured PowerPoint slide sets, image files such as Jpgs, Flash files or web site URLs, are auto-saved in the local project folder and/or auto-loaded into the online presentation folder as they are synchronised.
Thus even if a massive quantity of images, slide sets and so forth have been pre-loaded into the current project, only those used are resized to suit the chosen skin, then moved across - or up - to the project folder.
One great advantage of this method, is that skins can be changed at the last minute or after a presentation is archived.
All the slides and other assets are then auto-resized to fit the new skins' window structure. This can range from allowing a greater amount of space for the main slide region, to allowing more screen real-estate to the video window, adding additional regions, 'Email The Presenter' function and so forth.
The presentations can be auto-podcast if this is required.
The additional possibilities made available using RGB/VGA capture in
Accordent's Capture Station are discussed briefly below and again on the
Product Overview page.
...▲TOP...
Many presentations environments require an ability for the encoder to 'see' whatever device is plugged or switched into the room's data projector. In this method slide titles are relied on to generate bookmark points
if required. These can be fine tuned later if needed.
In Capture Station a presentation can be scheduled to auto-start at a certain time, auto-upload live in real time to a broadcast server then be available for archived streaming immediately following the presentation finishing. All with little or no input from the presenter.
The key to this is an additional input port in Capture Station supplied by a dedicated RGB/VGA card. While RGB screen capture has been around for a while, Accordent's 'Smart Technology' software takes it to a whole new level of sophistication and sensitivity. Read more about this HERE.
A quality look is all about lenses, light and resolution. Any old domestic video camera or mini web cam under normal room lighting can be used for capturing the video element of a Rich Media presentation.
However a modern DV camera (or two via Fire Wire switcher) with a good lens and some effort put into lighting the subject will make for a much more professional looking result. In the examples section, you will see presentations made both ways.
As we are only using the camera's lens and electronics input live to the PC via a DV (Fire Wire) cable, the tape format is not important.
The same camera 'front end' may have a Mini-DV, DV-Cam or HD-DV tape system on board. This makes no difference for our use capturing directly into the PC's encoder, running within Accordent's Presenter Plus,
Presenter Pro or Capture Station software environment.
There are some great new high-end domestic 3-chip cameras around. All you need worry about is the picture quality. We will not be using the camera's sound or tape format - which on these cameras is usually their weak point.
Setting the optimum capture resolution is done automatically in
Accordent encoders, based on the skin or template chosen pre-capture
and whether the output is to be 'High' 'Medium' or 'Low'. The main
thing to keep an eye on from an Australian point of view with all
conversion to streaming formats, is the default assumption by many streaming encoder products that they are for use with NTSC (read USA
& Japanese) cameras rather than Australian (and European) PAL
cameras. A little thought - and some once-only encoder profile
tweaks set up for you as part of an Australian installation - will
make these issues historical.
Successful web video encoding is all about using only what you must
have and throwing out the rest. If you can't tell the difference
between viewing 25fps (frames per second) encoded video and
12.5fps video at say 240x180, go for the latter. It will enable
other very visible quality factors to be improved for the same Kbps
bit rate. Or enable you to greatly reduce the bit rate for the same
file quality.
A similar rule goes for video size, both when framing the shot
with the camera and selecting frame size. The closer in you are to
your subject when capturing video, the more definition you will get
at the medium to low resolution end of the encoding scale. Look at
some of the differences on the Rich Media Demos
page.
A talking head is just fine in a 240x180 video window as part of a
Rich Media presentation. It is great in fact, as it allows a lot of
room for the region where the projector screen or synchronised slide
image appears. However a static wide shot of an entire lecture
theatre with presenter at the podium may need a larger video size
to be effective. This will reduce the screen real estate left for
the projector/slides region. In turn this will entail the presenter
needing to be more aware of small fonts and tiny images used in the
presentation.
An auto-tracking camera can make the process of keeping the
presenter framed at head and shoulders distance much easier,
although cheap models can make the viewer feel a bit seasick. The
camera pans, zooms and tilts following the presenter around the
stage at a pre-set distance using special sensors. New models are
being released regularly by major manufacturers. Look for the
quality of the lens and preferably a 3CCD imaging device.
Choose a template skin that as near as fills the monitor of the
majority of your audience as possible. It only takes a bit of
thought to have an 800x600 window template - because the General
Manager's laptop will only run at this resolution, running alongside
the same material presented in a 1200x700 version for everyone else.
This may be an unusual example, however it does happen. MS Internet
Explorer 7's zoom level control can help as well in this.
Try and avoid zooms, pans and fading in and out of things. Cuts only
from static locked-off shots is the way to go for a consistent, high
quality encoded result. One job we did recently was difficult to
encode well due to curtains flapping around uncontrollably behind
the subject. The encoder sees the curtain movement as equally
important to the subjects mouth movement. Both suffered due to the
visual information overload.
Encoders are very smart and treat a static background as requiring
little space to store at high quality. Even in a low resolution
streaming video file, you will notice a static background comes up
as very sharp. Even if at the same time the presenter's moving parts
look a little blurred...
...▲TOP...
■Great sound is more than a good microphone. A good pre capture card sound setup must be in place to ensure optimum level without distortion no matter what the presenter does. Even with an operator to adjust levels, it is not enough to plug a microphone straight into the PC input and be assured of it working every time.
The technology to use here, is a microphone preamplifier and limiter. This inputs a microphone level signal, converts it to line level signal and outputs to the PC via a limiter with gain compensation. Usually once set for an average voice level, it will not need adjusting no matter who is speaking. If a soft voice is coming in, the limiter does not work as hard. If a loud voice is on the mike, the limiter controls the maximum level permitted to the output. A huge dynamic range can be catered for in this way without sounding unnatural. More importantly, the recording level on the PC can be preset to just under the redline - and no matter what,
due to the limiter, it will not go beyond!
An additional advantage is that with the balanced line level signal coming out of the preamp/limiter, very long cable runs can be made with no loss in quality. In contrast, unbalanced or even balanced microphone signals can pick up noise from many sources including power cables and transformers that are always present in a hi-tech presentation room.
■The capture card needs to be the best. After experimenting with many capture cards and other PC input methods such as Fire Wire, I'm convinced a good dedicated card is the way to go.
Accordent recommend Osprey cards which give a terrific quality and will capture at high resolution on even a moderately paced PC. In addition they have balance line and microphone inputs, enabling perfect synchronisation of
the DV input video image and balanced analogue input limited line-level sound. After working like this once, nothing else will do...
...▲TOP...
While the power of Rich Media presentation delivery lies in its no-fuss 'this is what you would have seen if you were there' approach, behind the scenes things can be a little different.
For example, one of the most impressive features of the medium is the ability to add named bookmark index points during live capture of an event. In a seminar context where there are many hours being presented, if the named bookmarks are not done live it is near impossible to get them done later. The presenter is on a plane, the organisers are off to the Next Big Thing and so forth. In this context, the sheer joy experienced using Accordent's Presenter Plus for the first time can hardly be imagined.
The problem goes like this: Most presenters want to tell you about what they are going to say before they say it. This is great for preparing the audience for the gems to come. However important structural information can be passed on here that may be referred to later.
Often before the first headings slide is called up and usually before the diligent encoding assistant has even the faintest idea what a new topic's index marker could be named, the subject has been introduced. For best result, any named bookmark referring to a topic, needs to begin at its introduction. But how?
The speaker is now half way through point three.
The amazing folk at Accordent solved this little nightmare by allowing bookmark points to be dropped anytime - such as when you sense a topic change is imminent, then renamed and dragged around later. All this while the encoding and slide synchronising is still taking place!
If online delivery is post-presentation, index names and position can even be fine tuned after stopping the encoder. Or that night upon opening up the saved project. Simply put, there is no rendering involved here. The project file contains all the assets - slides, flash animations, URLs, bookmarks, video, gifs etc. - all of which can be changed and modified at will after the event without time consuming rendering.
So for example a typo-fix on a PowerPoint slide, a few index bookmark renames and an additional synchronised HTML reference, can be done within minutes. And then after you're sure you've finished, another typo, another index rename/replacement... all without slowing down a tightly scheduled post-presentation online delivery.
...▲TOP... |
