1 00:00:12,260 --> 00:00:16,980 This is unit 3B, Easy to understand and audio description. 2 00:00:16,980 --> 00:00:19,900 Element 3: Technical aspects. 3 00:00:20,500 --> 00:00:25,340 Video lecture: Easy to understand audio description principles, 4 00:00:25,340 --> 00:00:27,340 guidelines and conventions. 5 00:00:27,780 --> 00:00:29,100 This is Elisa Perego, 6 00:00:29,100 --> 00:00:31,600 from Università degli Studi di Trieste. 7 00:00:31,600 --> 00:00:36,940 In this video lecture I will tackle the technical aspects 8 00:00:36,940 --> 00:00:40,220 of the AD production process. 9 00:00:40,220 --> 00:00:44,600 I will also link them to the E2U context 10 00:00:44,600 --> 00:00:47,220 and focus on the aspects 11 00:00:47,220 --> 00:00:52,380 that can make an AD easier to listen to. 12 00:00:53,320 --> 00:00:57,700 Overall, the technical principles, aspects and conventions 13 00:00:57,700 --> 00:01:01,900 applied in standard AD production process 14 00:01:01,900 --> 00:01:07,300 apply to the production of Easy to Understand audio description. 15 00:01:07,300 --> 00:01:09,660 In short, these include: 16 00:01:10,140 --> 00:01:15,460 mastering AD software, or even subtitling software, 17 00:01:15,860 --> 00:01:18,780 which are used to generate time codes 18 00:01:18,780 --> 00:01:23,400 and to measure the time available for the AD between dialogues 19 00:01:23,400 --> 00:01:24,900 and sound effects. 20 00:01:25,300 --> 00:01:28,100 This is called spotting. 21 00:01:28,860 --> 00:01:32,940 Identifying the appropriate time code or a keyword 22 00:01:32,940 --> 00:01:37,940 when the voice talent reading your AD must start speaking. 23 00:01:38,860 --> 00:01:42,300 Providing a legend, at the beginning of your script 24 00:01:42,300 --> 00:01:45,140 where you define instructions. 25 00:01:45,140 --> 00:01:50,060 An F for fast, might invite the voice talent 26 00:01:50,060 --> 00:01:52,740 to speed up the reading peace a little. 27 00:01:53,400 --> 00:01:57,700 Noting when the voice can or can not go over the soundtrack 28 00:01:58,000 --> 00:02:04,000 or giving advice on the intonation depending on the source material. 29 00:02:04,620 --> 00:02:09,300 And finally, writing down in the AD script - in brackets - 30 00:02:09,300 --> 00:02:11,300 special pronunciation. 31 00:02:12,140 --> 00:02:16,340 These tips by Benecke and Volz, 32 00:02:16,340 --> 00:02:20,660 listed in the 2014 ADLAB guidelines, 33 00:02:20,660 --> 00:02:23,820 suggest that, in an E2U context, 34 00:02:24,870 --> 00:02:29,540 the AD script can be enriched with instructions 35 00:02:29,760 --> 00:02:35,600 that enhance the AD comprehensibility and listenability, 36 00:02:35,600 --> 00:02:38,980 be it a screen or an art AD. 37 00:02:39,890 --> 00:02:44,340 Exploiting a slower pace than in standard AD, 38 00:02:44,640 --> 00:02:50,220 emphasising words relevant to the comprehension of the AD 39 00:02:50,220 --> 00:02:51,780 and of the source text, 40 00:02:52,020 --> 00:02:55,740 clearly articulating long or complex words, 41 00:02:55,740 --> 00:02:57,740 and making frequent pauses, 42 00:02:57,970 --> 00:03:01,740 are simple tricks with a huge impact 43 00:03:01,740 --> 00:03:04,460 on the listenability of the final text. 44 00:03:06,700 --> 00:03:09,340 On a more technical level, 45 00:03:09,700 --> 00:03:15,600 Benecke and Volz point out that the finalized AD script 46 00:03:15,600 --> 00:03:19,300 is transformed from a written to an oral text. 47 00:03:20,020 --> 00:03:23,700 This implies the selection of the type of AD delivery: 48 00:03:24,300 --> 00:03:28,340 will it be human-voiced or a text-to-speech AD? 49 00:03:29,670 --> 00:03:34,980 The most appropriate voice talent should be selected, 50 00:03:35,330 --> 00:03:36,980 usually by genre. 51 00:03:38,000 --> 00:03:41,380 The recording is often done at a recording studio, 52 00:03:41,380 --> 00:03:45,140 where the sound director might be sight impaired 53 00:03:45,480 --> 00:03:48,180 or involve a sight impaired colleague 54 00:03:48,180 --> 00:03:53,700 to make user-centred decisions on intonation and speed. 55 00:03:54,860 --> 00:03:56,060 In this respect, 56 00:03:56,340 --> 00:03:59,820 to create an E2U AD, 57 00:03:59,930 --> 00:04:04,060 we could collaborate with an E2U user or expert 58 00:04:04,400 --> 00:04:07,860 who can validate the AD quality 59 00:04:07,900 --> 00:04:11,580 in terms of simplification and comprehensibility. 60 00:04:13,000 --> 00:04:15,740 Team work is in fact a good practice 61 00:04:15,740 --> 00:04:20,660 in both the AD and the E2U contexts, 62 00:04:21,100 --> 00:04:26,500 where the opinion of users and their feedback is a backbone. 63 00:04:27,920 --> 00:04:30,100 After the recording session, 64 00:04:30,100 --> 00:04:35,540 disturbing noises will be cleared by a sound technician or designer 65 00:04:35,540 --> 00:04:39,460 before mixing the AD with the original soundtrack 66 00:04:39,460 --> 00:04:41,720 in the case of audiovisual products. 67 00:04:43,150 --> 00:04:48,500 In E2U AD, optimising the reception of AD 68 00:04:48,500 --> 00:04:51,540 to make it easy-to-listen is paramount. 69 00:04:53,010 --> 00:04:57,340 In their chapter entitled: "Easier audio description", 70 00:04:57,340 --> 00:05:02,140 included in the volume edited by Sabine Braun ad Kim Starr, 71 00:05:02,140 --> 00:05:04,900 “Innovation in audio description research”, 72 00:05:04,900 --> 00:05:10,340 Bernabé and Orero offer food-for-thought on the subject. 73 00:05:11,000 --> 00:05:16,900 The authors insist on intervening on the technical aspects of AD delivery 74 00:05:16,900 --> 00:05:18,900 to improve its reception. 75 00:05:19,370 --> 00:05:25,000 They claim that interventions on the original version 76 00:05:25,000 --> 00:05:26,540 can improve it. 77 00:05:26,540 --> 00:05:30,620 For instance, by reducing noise and enhancing speech, 78 00:05:30,970 --> 00:05:35,650 relying on audio engeneering technology to highlight dialogue, 79 00:05:35,650 --> 00:05:40,380 and improve its intelligibility, especially if relevant. 80 00:05:40,380 --> 00:05:43,180 Or by modifying sounds 81 00:05:43,180 --> 00:05:47,580 with an aesthetic rather than diegenetic function, 82 00:05:47,580 --> 00:05:52,220 that interfere with dialogue accessibility. 83 00:05:53,820 --> 00:05:56,620 Reading speed is another key factor 84 00:05:56,620 --> 00:05:59,260 in easy-to-listen products. 85 00:05:59,850 --> 00:06:03,940 Setting a unique delivery rate is impossible 86 00:06:03,940 --> 00:06:09,220 given the diversified skills of humans in processing speech. 87 00:06:09,730 --> 00:06:14,970 Studies on the subject point to the need to diminish 88 00:06:15,170 --> 00:06:18,300 the number of words per minute 89 00:06:18,300 --> 00:06:21,580 to ensure improved listening comprehension. 90 00:06:22,160 --> 00:06:24,050 So, as a rule of thumb, 91 00:06:24,320 --> 00:06:28,500 E2U AD might consider avoiding 92 00:06:28,970 --> 00:06:35,000 the standard 180 or even 170 words per minute. 93 00:06:36,780 --> 00:06:39,780 Finally, it would be useful to enable users 94 00:06:39,780 --> 00:06:43,260 to personalize their E2U ADs 95 00:06:43,260 --> 00:06:48,620 and to choose preferred settings according to personal requirements 96 00:06:48,620 --> 00:06:50,620 or the situation. 97 00:06:51,320 --> 00:06:57,000 They could set a different AD volume according to their needs 98 00:06:57,000 --> 00:07:02,300 or to whether they are listening to ADs their smartphones 99 00:07:02,300 --> 00:07:03,500 or at the cinema, 100 00:07:03,500 --> 00:07:06,300 at home or at an open event. 101 00:07:07,550 --> 00:07:12,980 ADs are used in a variety of settings, 102 00:07:12,980 --> 00:07:17,540 and they cater for the accessibility of a variety of products: 103 00:07:17,590 --> 00:07:20,620 screen products, live exhibitions, 104 00:07:20,810 --> 00:07:26,140 external environments like gardens, museums and galleries, 105 00:07:26,540 --> 00:07:30,620 opera, theatre, dance, concerts, and many more. 106 00:07:32,200 --> 00:07:37,240 It is up to you to mix and match the recommendations you have learned 107 00:07:37,240 --> 00:07:38,820 find new ones, 108 00:07:38,820 --> 00:07:43,860 and determine if their efficacy depends on what you will be describing 109 00:07:43,860 --> 00:07:47,380 with comprehensibility and listenability in mind. 110 00:07:49,590 --> 00:07:52,300 All the pictures used in this presentation 111 00:07:52,390 --> 00:07:55,420 come from the Public Domain Vectors website, 112 00:07:55,420 --> 00:07:59,420 which gathers hundreds of royalty-free vector image 113 00:07:59,420 --> 00:08:05,340 under the Creative Commons Universal Public Domain Dedication license. 114 00:08:06,190 --> 00:08:09,620 This video lecture was prepared by Elisa Perego 115 00:08:09,620 --> 00:08:12,740 and produced by Andreea Deleanu, 116 00:08:12,740 --> 00:08:15,620 from Università degli Studi di Trieste. 117 00:08:16,920 --> 00:08:22,660 You can reach me at eperego@units.it.