1
00:00:12,700 --> 00:00:16,591
An agent made this explainer from
one paper, in about five minutes:

2
00:00:16,902 --> 00:00:20,351
the script, the slides,
the captions, and the video.

3
00:00:20,781 --> 00:00:23,247
Every sentence points
to the page it came from.

4
00:00:23,638 --> 00:00:26,825
And the paid voice? The
agent never saw the key.

5
00:00:29,200 --> 00:00:32,250
AI for Research Efficiency.

6
00:00:32,300 --> 00:00:35,610
Episode 12: Make a narrated
explainer of your paper.

7
00:00:38,700 --> 00:00:41,946
Last time, you mapped a
research field with OpenAlex.

8
00:00:42,485 --> 00:00:47,035
This time, one paper becomes a short
narrated video, for a lab's page,

9
00:00:47,085 --> 00:00:52,254
a class, or a grant's broader
impacts. One real run, replayed.

10
00:00:55,700 --> 00:00:57,382
The folder: the paper,

11
00:00:57,512 --> 00:01:00,326
an instruction file,
and three empty folders,

12
00:01:00,476 --> 00:01:03,420
for the work, the voice,
and the finished video.

13
00:01:04,130 --> 00:01:07,398
The paper is a 2025 review

14
00:01:07,448 --> 00:01:11,347
of water management in rice,
licensed CC BY:

15
00:01:11,397 --> 00:01:14,984
anyone may adapt it, with credit.
It's the review behind episodes 3 and 4.

16
00:01:15,834 --> 00:01:17,837
The instruction file sets the rules.

17
00:01:18,248 --> 00:01:22,806
About 150 words, one sentence per
line, each ending with its page.

18
00:01:23,237 --> 00:01:26,682
Only what the paper says,
and every number as printed.

19
00:01:27,512 --> 00:01:32,501
And one rule for the voice: the agent
has no key, and must not look for one.

20
00:01:32,933 --> 00:01:35,417
It writes a request,
and waits for the audio.

21
00:01:38,900 --> 00:01:42,385
The prompt asks for the script first,
checked against the paper;

22
00:01:42,776 --> 00:01:45,921
then the voice request,
the slides, and the video.

23
00:01:46,611 --> 00:01:51,394
In its first 10 seconds, the agent splits
the paper into one text file per page,

24
00:01:51,605 --> 00:01:53,906
so each sentence can carry its page.

25
00:01:54,825 --> 00:01:57,287
35 seconds in, a first draft:

26
00:01:57,597 --> 00:02:01,660
215 words, too long
for a minute. It cuts.

27
00:02:02,570 --> 00:02:03,572
Two more drafts.

28
00:02:03,983 --> 00:02:06,619
Then it checks each figure
against the page it cites,

29
00:02:06,710 --> 00:02:08,533
and writes down every check.

30
00:02:09,423 --> 00:02:12,470
At 82 seconds: 159 words

31
00:02:12,520 --> 00:02:15,466
in 13 sentences, each with its page.

32
00:02:17,300 --> 00:02:22,620
Methane fell by 31 to 62%:
page 1, in the abstract.

33
00:02:23,737 --> 00:02:25,760
Both gases together,
33.6 to 56.2% lower:

34
00:02:25,810 --> 00:02:29,836
page 3.

35
00:02:30,907 --> 00:02:34,813
Safe AWD raised yield by 8.9%,
over a 5-year experiment:

36
00:02:34,863 --> 00:02:37,546
page 5.

37
00:02:38,956 --> 00:02:43,220
Page 5 says one thing more:
the experiment was at a single site.

38
00:02:43,591 --> 00:02:46,994
The first draft said so;
cutting for length took it out.

39
00:02:48,270 --> 00:02:50,792
The numbers are right,
but the context is thinner.

40
00:02:51,223 --> 00:02:53,095
That's the check only you can make:

41
00:02:53,445 --> 00:02:56,347
does each sentence still
mean what the paper means?

42
00:02:58,100 --> 00:02:59,001
Now the voice.

43
00:02:59,471 --> 00:03:03,825
The agent writes a request: the text,
a stock voice, and the model.

44
00:03:04,215 --> 00:03:05,196
Then it waits.

45
00:03:05,961 --> 00:03:09,406
You make the paid request yourself,
outside the agent's session,

46
00:03:09,536 --> 00:03:14,863
with the key from your own .env file. The
companion page has a short script for it.

47
00:03:15,699 --> 00:03:21,038
First an estimate:
966 characters, about 107 credits.

48
00:03:22,151 --> 00:03:24,859
About 20 seconds later,
the audio is back,

49
00:03:25,070 --> 00:03:30,124
with the start and end of every
character. It cost 106 credits.

50
00:03:30,843 --> 00:03:34,954
The agent never saw the key.
It saw two files appear.

51
00:03:38,900 --> 00:03:41,985
Meanwhile,
the agent drew 12 slides in Python:

52
00:03:42,356 --> 00:03:45,221
one idea each, the key number large.

53
00:03:45,942 --> 00:03:48,589
It looked at them all at once,
and fixed what it saw:

54
00:03:49,041 --> 00:03:54,015
a clipped number, a drawing over a title,
and text running off two slides.

55
00:03:54,926 --> 00:04:00,477
The character timings become captions:
a sentence at a time, at most two lines.

56
00:04:01,348 --> 00:04:06,462
Then ffmpeg puts it together: each
slide while its sentences are spoken,

57
00:04:06,713 --> 00:04:09,159
the voice,
and the captions burned in.

58
00:04:10,050 --> 00:04:14,555
It checks the result: the length
matches the voice, every slide appears,

59
00:04:14,785 --> 00:04:19,070
and the captions can be read. It made
their boxes darker until they could.

60
00:04:19,756 --> 00:04:22,966
5 minutes and 7 seconds,
start to finish.

61
00:04:26,900 --> 00:04:31,545
Its report is candid:
the video runs 89 seconds, not 60.

62
00:04:31,956 --> 00:04:36,620
This voice reads slowly, and the
paper's address alone takes 12 seconds.

63
00:04:37,566 --> 00:04:40,855
And it names what it could not check:
it cannot listen.

64
00:04:41,266 --> 00:04:45,557
So listen yourself; every number should
sound the way the paper prints it.

65
00:04:46,567 --> 00:04:49,790
Before you share it:
each sentence against its page,

66
00:04:49,960 --> 00:04:52,993
the voice against the numbers,
the credit to the paper,

67
00:04:53,283 --> 00:04:56,105
and a line that says the voice is AI.

68
00:04:58,100 --> 00:04:59,441
About the voice service.

69
00:04:59,871 --> 00:05:02,583
ElevenLabs' free plan
can't be used commercially,

70
00:05:02,753 --> 00:05:07,516
and anything you publish from it
must have elevenlabs.io in its title.

71
00:05:08,181 --> 00:05:10,364
Paid plans include a
commercial license,

72
00:05:10,454 --> 00:05:14,259
and what you make during a subscription
stays licensed after it ends.

73
00:05:14,970 --> 00:05:17,434
Never clone someone's voice
without their consent,

74
00:05:17,644 --> 00:05:20,067
and never hide that a voice is AI.

75
00:05:21,073 --> 00:05:25,966
This series is narrated with ElevenLabs,
on a paid plan we use and value.

76
00:05:26,277 --> 00:05:30,046
A free voice that runs on your own
computer is on the companion page.

77
00:05:31,700 --> 00:05:34,111
Here is what the run made, in full.

78
00:07:07,700 --> 00:07:11,867
Try it with your own paper: copy the
instruction file from the companion page,

79
00:07:12,118 --> 00:07:13,450
ask for the script first,

80
00:07:13,701 --> 00:07:17,547
and check every sentence against
its page before you make the voice.

81
00:07:22,100 --> 00:07:26,103
So: pin every sentence to its page,
keep the key with you,

82
00:07:26,433 --> 00:07:30,136
listen before you share,
and say the voice is AI.

83
00:07:31,500 --> 00:07:34,078
Next: how this series is made.