Cisco Digital Video Recorder in Stand-Alone Mode Uživatelský manuál stáhnout pdf (Strana 21)

6 - Cisco Lecture Vision

Cisco Systems

System Description

Cisco Lecture Vision is a lecture recording solution which takes its

input from one or more Cisco video codecs. For the demo, the

recording was accomplished using a Telepresence Content Server

(TCS), which was located in a Cisco Data Centre, so very little

equipment was needed in the lecture theatre venue. Typically the

codec takes three inputs as follows: speaker audio, speaker video and

the presentation video. The codec then combines these into an audio-video stream, which is passed

to the TCS and recorded. During a live event, it is also possible for other codecs to join the call and

view and interact with the event live. At the end of the recording session, the recorded video is passed

to the Cisco Media Experience Engine where it is transcoded (for playback on various output devices)

and processed by Pulse Analytics to identify an audio profile for each speaker and to produce a key

word index based on the audio track content, matched against a pre-loaded vocabulary. The finished

video, along with the speaker index and key word index, is then published on Cisco’s Show and Share

video portal. It is further possible to manually add chapter markers if needed. Show and Share

supports a variety of playback formats including full screen and the output can be viewed on mobile

devices such as iPad or an android device. When viewing the recording, users can click on the key

word index to jump to the point(s) in the video where the indexed word is spoken. Pulse Analytics is

performed once the video has been recorded so there is a delay in time of approximately two times

the length of the original video before it can be viewed.

Screenshot

https://share.mxabu.com/vportal/VideoPlayer.jsp?ccsid=C-94cd5523-7f08-4c80-af52-241efe0c3384:1

Overview

Equipment used on site to achieve the recording was a Cisco Tandberg C60, which established a

H323/SIP connection to a TCS at the data centre in San Jose, United States (US). This central server

then recorded the video stream session. The recording process began by requesting Cisco Tandberg

C60 to call the TCS, which effectively required pressing one button. Once the recording call was

ended, the server automatically enriched the content with metadata identifying the speakers, and

transcribed the spoken words (speech-to-text). Everything was then indexed and published on a

website through a Flash application.

1 2 ... 16 17 18 19 20 21 22 23 24 25 26 ... 29 30

Žádné komentáře

Cisco Digital Video Recorder in Stand-Alone Mode Uživatelský manuál Strana 21