1 00:00:00,400 --> 00:00:02,680 This is an ADLAB PRO video. 2 00:00:02,760 --> 00:00:06,080 ADLAB PRO is a project funded by the European Union. 3 00:00:06,160 --> 00:00:10,000 This video is made available to you on a Creative Commons licence. 4 00:00:10,160 --> 00:00:14,000 Hello. I’m Pilar Orero, from Universitat Autònoma de Barcelona, 5 00:00:14,080 --> 00:00:18,280 and in this ADLAB PRO video I'll talk about technology delivery. 6 00:00:18,360 --> 00:00:20,840 This is Unit 2 in Module 6. 7 00:00:21,000 --> 00:00:24,560 Technology these days allows for the delivery of AD 8 00:00:24,640 --> 00:00:28,760 in a live situation or recorded. It also allows for the production 9 00:00:28,840 --> 00:00:32,640 to be in the cloud or local. Written by one person 10 00:00:32,720 --> 00:00:37,160 or by a collaborative group. You can work in situ or remotely. 11 00:00:37,240 --> 00:00:41,760 Finally, there are three ways to produce an AD: automatically, 12 00:00:41,840 --> 00:00:44,480 that is 100% automatic, by a human, 13 00:00:44,560 --> 00:00:47,880 or using a hybrid or semi-automatic model, 14 00:00:48,280 --> 00:00:52,920 that is, when a human post-edits an automatically-made AD. 15 00:00:53,080 --> 00:00:56,360 In fact, AD has two very different production cycles, 16 00:00:56,480 --> 00:00:58,480 which affects the delivery: 17 00:00:58,560 --> 00:01:02,160 the first is the production of the script, which is later voiced. 18 00:01:02,240 --> 00:01:05,560 The second is the delivery, which can be performed automatically 19 00:01:05,640 --> 00:01:08,480 through text-to-speech or can be read by a human. 20 00:01:08,680 --> 00:01:11,440 Regarding delivery and linked to the production, 21 00:01:11,600 --> 00:01:16,400 we have four situations. Recorded is when we have a movie with an AD, 22 00:01:16,560 --> 00:01:19,760 and this has been previously prepared and recorded, 23 00:01:19,840 --> 00:01:23,440 you can’t alter that. Also, there is no possible interaction 24 00:01:23,520 --> 00:01:25,680 because we are dealing with a finished product. 25 00:01:25,760 --> 00:01:28,840 This is not the case for an opera or a theatre play, 26 00:01:28,920 --> 00:01:31,720 where we can have a draft AD generated 27 00:01:31,840 --> 00:01:35,560 by attending the many rehearsals before the day of the event, 28 00:01:35,640 --> 00:01:39,240 but every day the music tempo and the actors may improvise. 29 00:01:39,320 --> 00:01:43,680 So the AD is delivered as prepared, but some adaptations 30 00:01:43,760 --> 00:01:45,920 may have to be introduced on real time. 31 00:01:46,040 --> 00:01:50,760 We have the AD of a tennis match, where we don’t have the result, 32 00:01:50,840 --> 00:01:54,440 and every minute of the match is a live situation and is improvised. 33 00:01:54,600 --> 00:01:59,000 And finally we have the AD of a TV soap opera, for example. 34 00:01:59,080 --> 00:02:01,680 Or a TV series which is produced and delivered 35 00:02:01,760 --> 00:02:05,400 by a team of audio describers and it takes place over many years 36 00:02:05,480 --> 00:02:09,320 or over many seasons. And for that you would need to have 37 00:02:09,400 --> 00:02:13,720 a style sheet and also to have a knowledge of the characters, 38 00:02:13,800 --> 00:02:17,160 the plot, and all the situations that took place in the past. 39 00:02:17,400 --> 00:02:19,920 How an AD is produced varies 40 00:02:20,000 --> 00:02:24,520 if we use an editor installed in our PC, or an editor in the cloud. 41 00:02:24,600 --> 00:02:27,560 You can also produce an AD without an editor, 42 00:02:27,680 --> 00:02:31,720 with a piece of paper and a pen, or using a text processing software 43 00:02:31,800 --> 00:02:35,280 in your PC and taking down the timeframes manually. 44 00:02:35,360 --> 00:02:38,520 And you can do it live over the microphone as it happens. 45 00:02:38,720 --> 00:02:42,560 You can also be present physically when delivering an AD, 46 00:02:42,640 --> 00:02:45,120 but you might be in a remote location, 47 00:02:45,200 --> 00:02:48,040 so actually you may be doing a football match, 48 00:02:48,120 --> 00:02:51,640 but not being on the field, you may be at home, for example. 49 00:02:51,720 --> 00:02:55,880 Artificial intelligence can be a great help, since increasingly 50 00:02:55,960 --> 00:02:59,920 it will provide added information by automatic locations 51 00:03:00,000 --> 00:03:03,040 or information about the context also automatically. 52 00:03:03,200 --> 00:03:06,320 The production of the text, apart from human-produced, 53 00:03:06,400 --> 00:03:09,200 may be through automatic visual tagging technology, 54 00:03:09,280 --> 00:03:13,320 or generated using the existing linguistic assets, for example, 55 00:03:13,400 --> 00:03:16,560 the script of a movie where it describes the set. 56 00:03:16,640 --> 00:03:21,440 Most automatically-produced ADs are later post-edited by a human. 57 00:03:21,640 --> 00:03:26,440 The sound production can be human, when a voice artist reads the AD, 58 00:03:26,520 --> 00:03:28,840 or based on text-to-speech technology, 59 00:03:28,920 --> 00:03:31,560 which also allows for voice choice. 60 00:03:31,640 --> 00:03:35,960 The speed of the delivery can be altered increasing or decreasing it. 61 00:03:39,120 --> 00:03:41,280 The preparation of this presentation 62 00:03:41,360 --> 00:03:44,960 was supported by ADLAB PRO. Audio Description: 63 00:03:45,040 --> 00:03:48,680 A Laboratory for the Development of a New Professional Profile 64 00:03:48,800 --> 00:03:53,080 financed by the European Union under the Erasmus+ Programme, 65 00:03:53,160 --> 00:03:56,440 Key Action 2, Strategic Partnerships. 66 00:03:56,520 --> 00:03:57,720 Project number: 67 00:03:57,800 --> 00:04:06,160 2016-1-IT02-KA203-024311. 68 00:04:06,720 --> 00:04:10,040 The information and views set out in this presentation 69 00:04:10,120 --> 00:04:13,160 are those of the authors and do not necessarily reflect 70 00:04:13,240 --> 00:04:16,320 the official opinion of the European Union. 71 00:04:16,400 --> 00:04:19,760 Neither the European Union institutions and bodies 72 00:04:19,840 --> 00:04:22,360 nor any person acting on their behalf 73 00:04:22,440 --> 00:04:24,280 may be held responsible for the use 74 00:04:24,360 --> 00:04:27,000 which may be made of the information contained therein. 75 00:04:27,200 --> 00:04:29,200 This was an ADLAB PRO video. 76 00:04:29,280 --> 00:04:32,920 ADLAB PRO partners are: University of Trieste, 77 00:04:33,000 --> 00:04:36,000 Universitat Autònoma de Barcelona, 78 00:04:36,080 --> 00:04:40,080 University of Antwerp, Adam Mickiewicz University, 79 00:04:40,160 --> 00:04:43,160 RTV Slovenia, Utopian Voices, 80 00:04:43,240 --> 00:04:47,440 Soundfocus and Royal National Institute of Blind People. 81 00:04:47,520 --> 00:04:50,480 ADLAB PRO was funded by the European Union. 82 00:04:50,560 --> 00:04:54,320 This video was made available to you on a Creative Commons licence.