Recording a meeting
The context for most recording or broadcasting activity is a meeting or event at which an expert speaker presents material which has value for a wider audience, or as reference material. Whilst it has been customary for each speaker to provide a copy of their slides, is often difficult to reconstruct the speaker's argument, or recall the finer points, without a record of their words. For this reason, audio is valuable and all recordings should include an audio track. In some circumstances, audio plus slides may provide a sufficient record. However, a survey of one member group suggested a strong desire for video content as well.
Most speakers will use electronic slides (and possibly other electronic material) as visual aids to support their presentation. It's important to capture these as well as the audio/video stream, but experience to date suggests that video capture is not an ideal way of doing this: too often, a video camera is used to capture the speaker's face at the expense of the slides or a poor quality copy of the slides. Therefore, a PDF file or direct “screen grab” of the electronic material must be captured as well. Once captured, it is often desirable to synchronise the slides with the audio/video stream and merge them into a single stream for publication. The offline and online scenarios differ in the way in which this is achieved.
The differences between scenarios therefore lie in the nature of the material captured (either audio+slides or audio+video+slides), the equipment used, the file size generated, and the effort required. Recordings will often be followed by an editing step, during which the recording is improved and converted into an appropriate format for publication. The scenarios may also be combined in various ways to achieve a given group's goals. For example, it may be desirable to have a live broadcast of an event at moderate quality and subsequently produce a higher quality, edited version for longer term reference.
The basic idea behind the Recording & Broadcasting scenarios is that they can act as simple recipes which any member group, even one with a low budget and limited skills, can adopt in a straightforward way. They have already been tested and used for routine events, so any group adopting one of them can be reasonably confident that it will work. The scenarios are also graded so that skills developed using the early scenarios can be re-used and enhanced with later scenarios.
Nine recording scenarios are defined:
Offline SD Video
Online SD Video
SD Video Live Broadcasting
Offline HD Video
Online HD Video
HD Video Live Broadcasting
In all cases, “audio” implies that both slides and audio will be captured and “video” implies that slides, audio and video will be captured.
Scenario 0: Documentation only
This represents existing good practice where a record of the meeting is provided in the form of a copy of the presenter's slides, or a transcript of what was said, or both. These documents should be provided in PDF form because that is most widely used and is the most convenient for downloading over the web. The advantages of this scenario are that documentation is compact, easily transmitted over networks, and requires no additional software or skills. The disadvantages are that the presenter's slides seldom provide a complete record of the meeting, and that transcripts are expensive or time-consuming to produce.
Scenario 1: Offline Audio
This scenario uses an inexpensive digital voice recorder, sometimes equipped with a tie clip microphone, to record an audio stream to a storage device in the recorder. The speaker carries the voice recorder in his/her pocket. Typical consumer recording devices use an SD card for storage and have a capacity of around 30 minutes of audio. [For very short recordings, a smart phone could substitute for the voice recorder but this is not recommended for larger/longer meetings.] Semi-professional devices may be capable of capturing much longer recordings and/or higher quality. There is often no control of the recording device while it's being used, although some devices may enable monitoring of sound levels.
Data from the storage card must imported into a computer, usually as an MP3 file, which may be used directly or edited. Speaker slides must be captured separately, usually as a PDF file, and there is generally no synchronisation between audio and slides. They may be manually synchronised for web publication via SlideShare.
Scenario 2: Online Audio
This scenario uses a microphone connected to a laptop computer, equipped with recording software such as Audacity, to capture an audio stream directly to the computer. Recent experience with remote microphones, including directional microphones, suggests that sound quality may be unreliable unless the microphone is actively managed. Therefore, a tie clip microphone is normally a better option, connected to the laptop via a long cable or a wireless microphone kit, using an audio jack or USB connector. Some speakers like to move around and are unhappy with the idea of being tied to a cable, so a wireless microphone is the recommended solution.
The wireless kit consists of the microphone plus a small battery powered transmitter which is worn by the speaker, and a base station receiver which is connected to the laptop. An advantage of a wireless kit is that the receiver and laptop may be positioned remotely from the speaker. It is often convenient to operate this configuration from the back of the lecture room, where a volunteer can start and stop the recording, monitor sound levels, switch microphones, etc without disturbing the speaker.
Normally, the software used for recording is also used for subsequent editing. With a suitable hard disk drive installed in the laptop, or an external hard drive, recordings of one hour up to several hours may be captured easily. The best known software for this purpose is the free, open source Audacity which may be downloaded from http://audacity.sourceforge.net/download/. Other audio recording software may also be used, but will usually be priced. (Alternatively, both slides and audio track may be captured using Camtasia, by selecting no camera input.)
Audacity provides a convenient “dashboard” style user interface for operating the recording, plus a waveform display used during recording and editing. It stores recordings using the .aup file format and supports editing in this format, although a faster computer may be required for editing. There is a fairly comprehensive tutorial at http://wiki.audacityteam.org/wiki/Audacity_Wiki_Home_Page. The edited recording can be converted to MP3 format for distribution using an additional software package known as Lame. Speaker slides must be captured separately as a PDF file because Audacity does not support capture of this data. There is no synchronisation between slides and audio, and two files must be published to capture the complete recording. However, when a recording is published on the web using the SlideShare service, slides can be manually synchronised with the audio track.
Scenario 3: Offline Video at Standard Definition
This scenario uses a digital camcorder plus microphone to record a combined audio + video stream to a storage device such as an SD card or a hard disk drive. Normally, the camcorder should be mounted on a tripod somewhere towards the back of the lecture room where it will not disturb the speaker and should have a long enough focal length lens to make this possible. A number of popular makes of digital camcorders are currently available with wide range zoom lenses, which makes it possible to select a convenient focal length. Most camcorders have a built-in directional microphone but sound quality may be unreliable and therefore use of an external microphone is recommended. This could be a camera-mounted microphone or a wireless microphone of the type described in Scenario 2. In either case, it's important that the camcorder have an input socket for the external microphone, to enable both audio and video tracks to be recorded on the storage device.
Standard Definition (SD) video uses a resolution roughly equivalent to that used in conventional analogue TV pictures or VGA computer monitors, and often defined as 720 x 576 pixels. This enables long recordings to be captured on a single storage card although the length of recording also will depend on the storage capacity of the card. SD cards are currently available with capacities up to 32 Gbytes, which should allow several hours recording.
The camcorder and microphone combination is usually operated by a volunteer during the meeting. The combined audio + video tracks on the storage device must then be imported into a computer running a suitable video editor such as Camtasia, which is also capable of capturing speaker slides (or other displayed information). Synchronisation between the slides and audio/video track may be established during editing. The editing recording willl normally be rendered into MPeg4 format for publication on the web.
Scenario 4: Online Video at Standard Definition
This scenario uses video camera and separate microphone, both of which are connected to a laptop computer equipped with recording software, to capture audio and video streams directly to the computer. One advantage of this approach is that the recording software can also capture the speaker slides via “screen grab” technology and synchronise it with audio/video stream during the recording. Alternatively, the camcorder may be equipped with an external microphone as described on Scenario 3.
The choice of cameras which may be connected to a laptop includes external webcams and camcorders attached via a digital bus, such as Firewire, or an analogue to digital conversion device, often known as a USB video grabber. Some higher priced video cameras have the Firewire feature, but we were unable to find it on consumer models. However, most camcorders provide an analogue output signal which may be converted to an SD quality digital signal via a conversion device. These are readily available at around £20 and we recommend the LogiLink USB video grabber. This is supplied with drivers for Windows computers and free open source driver is available for Mac computers from www.bentrask.com.
Most external webcams have a short focal length lens and must be placed close to the speaker. This requires a USB extension cable for connection to the laptop, since it is normally convenient to locate this and other equipment at the back of the room where a volunteer can start and stop the recording, manage camera angle, recording levels, etc. Camcorders usually have wide range zoom lens and may be placed at the back of the room, close to the laptop and other equipment.
To make use of the “screen grab” function in recording software such as Camtasia, the speaker's slides must be installed on the same laptop which is being used for the audio/video recording. It is therefore necessary to copy the speaker's slides to the recording laptop before the session begins. The person operating the laptop should be alert to following the speaker's slide changes as they occur.
Normally, the software used for recording is also used for subsequent editing. File sizes will typically be an order of magnitude larger than with audio recording but a modern laptop with a suitable hard disk drive, or a high capacity external hard drive, should enable recordings of several hours to be captured easily. Experience suggests that Camtasia is suitable for this purpose. Its “screen grab” function uses a rectangular marquee to denote the area of the screen to be captured and this would normally be positioned over the window used for the speaker slides, but could capture the output from any application. The benefit of performing this screen grab in realtime is that accurate synchronisation may be achieved during the recording process, avoiding work during the subsequent editing process.
Camtasia is available from Techsmith Labs, together with other related products. See http://shop.techsmith.com/store Action=DisplayHomePage&Currency=GBP&Locale=en_GB&SiteID=techsmit for details. It provides a convenient “dashboard” style user interface for operating the recording, plus a waveform display similar to that provided by Audacity for the audio stream, which is used during recording and editing. It records composite files in its own format and supports editing in this format, although a faster computer may be required for editing. As previously noted, speaker slides may be incorporated into the video output stream and synchronisation established during the editing process.
Scenario 5: Live Video Broadcasting at Standard Definition
This scenario uses the Tricaster system to both record and broadcast standard quality video. A full description is beyond the scope of this report; users should refer to the Tricaster manual and undertake specific training. Tricaster can only be used in the Wilkes 1 and 2 rooms in the Davison Building.
Tricaster is a semi-professional recording and broadcasting system incorporating a camera control console, a video mixing desk, and a personal computer. The console enables individual control of the meeting room cameras which may be panned or zoomed via a joystick control. It also allows pre-set camera positions to be programmed and selected at the press of a button, so that the meeting coordinator can select a particular view quickly. The video mixing desk enables the coordinator to view the output of all cameras and select in real time which will be used for output to a recording device or broadcast channel. In addition to cameras, the mixing desk can be used to select pre-recorded still pictures, e.g. BCS logo, meeting title, etc., and external video feeds from other sources such as the data projector which is showing the speaker's slides. The mixing function includes transition effects such as fade in/fade out. The PC is used to host recording software and manage the resulting files. Video output can be forwarded to an external streaming server by running Adobe Flash Live Media Encoder, which will encode the video stream in Flash format and allows specification of a URL where the server can be found.
In the hands of a skilled operator, Tricaster enables real time selection of which inputs should be recorded or broadcast, thereby avoiding work which would otherwise be required during the editing stage. However, it also requires significant preparation before each event to set up camera positions, meeting titles, video feeds, etc., and processing after the event to render recorded data into a compressed form suitable for long term storage. Video data is recorded at SD quality because that is the capability of the meeting room cameras and the iVGA feed from the data projector.
Scenario 6: Offline Video at High Definition
This scenario is very similar to Scenario 3 because most modern consumer camcorders have HD capability. When recording in HD mode, they will capture video at a resolution of either 1280 x 720 pixels or 1920 x 1080 pixels (“full HD”). At 1280 x 720 resolution, approximately two hours recording can be captured on a single 16 Gbyte storage card, or approximately four hours on a 32 Gbyte card. At 1920 x 1080 resolution, recording times are less than half these numbers. Some modern digital cameras, particularly digital single lens reflex cameras, have an HD video recording mode which also records to an SD memory card. Whilst these cameras have a wide choice of superior lenses which can be used with them, in practice they are less useful for video work because of limitations in storage capacity or maximum recording time, which may be as little as 20 minutes.
The camcorder and microphone combination is usually operated by a volunteer during the meeting. The combined audio + video tracks on the storage device must then be imported into a computer running a suitable video editor such as Camtasia. The editing and publication process follows the same approach as used in Scenario 4, but note that Camtasia is also capable of capturing speaker slides (or other displayed information) so that synchronisation between the slides and audio/video track may be established during editing. Even at HD resolution, direct video capture of the meeting room projector screen is usually inadequate and therefore it's necessary to include of copies the speaker's slides during the editing process, in the manner described for Scenario 3.
Scenario 7: Online Video at High Definition
This scenario is similar to Scenario 4, using High Definition capture devices in place of Standard Definition devices. However, these include HD webcams, especially tripod mounted studio webcams which are now available at low prices, or HD camcorders. These offer an especially cheap and easy-to-use form of digital input for recording.
Unfortunately, consumer HD camcorders provide digital output only in the form of an HDMI connection, which can be connected to a device such as digital TV or Blu-Ray recorder but not to most laptops. Few conversion devices are available to provide the needed connection to a laptop, and these are typically high priced. Another option may be a semi-professional video camera with a Firewire connection. These provide high quality digital input to a laptop but are mostly very high priced. Second hand cameras with this capability are sometimes available more cheaply on e-Bay. The use of cheap analogue-to-digital conversion devices with analogue output from the camcorder reduces quality to SD level, and is therefore not recommended.
Normally, the software used for recording is also used for subsequent editing. File sizes will typically be an order of magnitude larger than with audio recording but a modern laptop with a suitable hard disk drive, or a high capacity external hard drive, should enable recordings of several hours to be captured easily. Experience suggests that Camtasia is suitable for this purpose. Its “screen grab” function uses a rectangular marquee to denote the area of the screen to be captured and this would normally be positioned over the window used for the speaker slides, but could capture the output from any application. The benefit of performing this screen grab in realtime is that accurate synchronisation may be achieved while avoiding work may during the subsequent editing process. It's worth bearing in mind that a number of web video publishing sites, such as YouTube, will apply high levels of compression to video content when it is published.
Scenario 8: Live Video Broadcasting at High Definition
The combination of HD video capture devices, such as webcams built in to modern laptops, and online meeting software enables high quality videoconferencing between small groups of participants. A wide range of products are available which support this capability, some of which are free, and some for which a charge is payable. Conferencing with over 16 users is normally charged at a higher rate than for less smaller numbers.
This type of software is widely used to support webinars with moderate audience sizes but the assumption is that the presenter is seated at a desk. However, it can be adapted to use in lecture style meetings by using a tripod mounted studio style webcam which provides a view of the speaker and the area immediately surrounding him/her. Audience numbers will be limited to that supported in an online meeting.
Uniquely, Google+ offers both an online meeting capability, known as "Hangout", and an “on air” option which enables video data from the meeting to be broadcast live to a related YouTube account. This provides a live but non-interactive broadcast of the meeting which is very appealing to some audiences. Audience numbers are limited only by what YouTube can support, so are unlimited for practical purposes.
Another attractive feature of this offering is that the software automatically records the video stream and posts the completed recording on YouTube after the event has finished. This avoids post-event editing work although YouTube compresses the video stream and reduces quality to approximately SD resolution. A sensible approach may be to make a higher quality recording in parallel (using the technique described in Scenario 3), which can be edited and posted on the web at a later time.
At least two participants must be available to set up the online meeting, before it can be broadcast. These participants must have Google+ accounts but those viewing the broadcast do not need Google+ or YouTube accounts. The online meeting is established by the meeting coordinator who should initiate a Hangout and invite at least one other participant, preferably a volunteer using a nearby laptop. Live broadcasting is triggered by selecting “Hangout on Air” while setting up the online meeting, which causes the screen data to be transmitted to the linked YouTube account where it will appear as video. To view the broadcast, participants merely have to search for the meeting coordinator's channel or find it via an e-mail link.
By default, the Hangout will use the built-in webcam and microphone in the coordinator's laptop, and those of other participants. It's therefore necessary to select the right camera using the Settings panel and it's usually a good idea to mute these other cameras and microphones, to avoid an interrrupted dialogue, distracting pictures, background noise or avoid feedback occurring. In practice, it is essential to monitor both the output from the online meeting and the YouTube broadcast to confirm that video and audio quality is acceptable, which may require use of three laptops.
Video input of the speaker may be captured from a cable connected webcam placed nearby, or from a camcorder attached via a video grabber device, as described in earlier scenarios. Audio input may be captured from a webcam. This is usually satifactory when the speaker's voice is amplified by meeting room speakers but otherwise may appear distant. Alternatively, a wireless microphone worn by the speaker may be used, with the base station connected to the meeting coordinator's laptop. These settings can be changed during the course of the hangout, enabling the use of close up, wide angle, and other camera views. Speaker slides or demonstrations may also be shown by selecting the “share screen” option, but note that a copy of the slides must already be installed on the coordinator's laptop.
Google+ also supports “applications” such as text chat which may be run within the Hangout window. This is often useful for side coversations between those operating the Hangout.