1
00:00:00,160 --> 00:00:12,400
...

2
00:00:12,720 --> 00:00:14,480
In your opinion,
if such a tool were available,

3
00:00:14,840 --> 00:00:16,120
how would it make it easier for you

4
00:00:16,440 --> 00:00:18,440
to access more information
or audiovisual content?

5
00:00:20,760 --> 00:00:23,840
- About the signing avatar project,

6
00:00:24,160 --> 00:00:28,880
about the news
and all the audiovisual content,

7
00:00:29,200 --> 00:00:32,000
it seems pretty important to me.

8
00:00:32,360 --> 00:00:35,080
We're in a world
based on vocal languages,

9
00:00:35,400 --> 00:00:39,680
and automatic translation
allows to communicate fairly easily,

10
00:00:40,000 --> 00:00:42,600
but deaf people don't have such a tool.

11
00:00:42,960 --> 00:00:47,880
The existence of such a tool
can make things a lot better,

12
00:00:48,200 --> 00:00:51,680
and I think it's going to open new doors

13
00:00:52,000 --> 00:00:53,160
for the deaf.

14
00:00:53,480 --> 00:00:56,840
Do you think the communities of deaf
and hard-of-hearing people

15
00:00:57,160 --> 00:01:00,040
would be in favor of such a tool?
Could you explain?

16
00:01:02,360 --> 00:01:06,840
As for a signing avatar
for deaf people, hard of hearing people,

17
00:01:07,160 --> 00:01:08,320
hearing people,

18
00:01:08,640 --> 00:01:09,960
I really wonder

19
00:01:10,280 --> 00:01:14,720
whether they're going to find it positive
or not.

20
00:01:15,040 --> 00:01:17,360
There will be those for and those against.

21
00:01:17,680 --> 00:01:21,760
Some people will not want
to follow such technological evolution,

22
00:01:22,080 --> 00:01:25,480
and they'll always prefer
to call on an interpreter,

23
00:01:25,800 --> 00:01:26,920
for good reasons.

24
00:01:27,240 --> 00:01:31,960
But it will allow evolution
in the field of translation,

25
00:01:32,280 --> 00:01:36,400
and for me, it might be able
to address some situations

26
00:01:36,720 --> 00:01:39,280
that an interpreter can't answer.

27
00:01:39,600 --> 00:01:42,800
So I think there will be positive
as negative aspects.

28
00:01:43,120 --> 00:01:46,720
You will just have to take the time
that people get used to

29
00:01:47,040 --> 00:01:48,000
this progress.

30
00:01:48,320 --> 00:01:52,600
To what extent do you think
a virtual signer for audiovisual programs

31
00:01:52,920 --> 00:01:55,840
(especially journalistic content)
can help you?

32
00:01:56,160 --> 00:01:57,960
Personally? Professionally?

33
00:02:01,280 --> 00:02:03,520
So... I work at Media'Pi,

34
00:02:03,840 --> 00:02:07,320
which is a sign language news website.

35
00:02:07,640 --> 00:02:11,680
And it's true that sometimes we're stuck
because our journalists

36
00:02:12,000 --> 00:02:14,760
have trouble accessing some information.

37
00:02:15,080 --> 00:02:17,400
So they might need this kind of tool.

38
00:02:17,720 --> 00:02:21,920
It also happens that they want
to participate in conferences

39
00:02:22,240 --> 00:02:25,280
without the means to hire an interpreter.

40
00:02:25,600 --> 00:02:29,920
So in terms of accessibility,
it could be useful to them.

41
00:02:30,240 --> 00:02:33,760
And for everything
about urgent news,

42
00:02:34,080 --> 00:02:38,040
which in the hearing world
are circulating pretty fast,

43
00:02:38,360 --> 00:02:42,160
we could actually broadcast information faster

44
00:02:42,480 --> 00:02:46,000
to deaf people through this signing avatar.

45
00:02:47,000 --> 00:02:50,960
For me, I think it's just benefits.

46
00:02:51,280 --> 00:02:52,800
It's going to give us

47
00:02:53,120 --> 00:02:56,280
a greater opening, a greater accessibility.

48
00:02:56,600 --> 00:02:59,440
What would you expect from this project?

49
00:03:02,760 --> 00:03:05,080
As regards this project,

50
00:03:05,400 --> 00:03:08,480
my expectations are first
professional.

51
00:03:08,800 --> 00:03:10,440
Working in Media'Pi,

52
00:03:10,760 --> 00:03:15,600
I'm thinking, for example,
accessing information immediately

53
00:03:15,920 --> 00:03:19,280
without having to go through journalists,

54
00:03:19,600 --> 00:03:22,920
through sign language translation, etc.,

55
00:03:23,240 --> 00:03:28,080
being able to broadcast live information
through this system

56
00:03:28,400 --> 00:03:31,800
would allow us to save time
and inform deaf people

57
00:03:32,120 --> 00:03:34,840
at the same speed as the hearing people.

58
00:03:35,160 --> 00:03:39,320
And then in the comments we have,

59
00:03:39,640 --> 00:03:46,240
we often have things
that are expressed anonymously.

60
00:03:46,560 --> 00:03:49,720
And deaf people don't have that anonymity

61
00:03:50,040 --> 00:03:53,160
when they want to communicate
or comment.

62
00:03:53,480 --> 00:03:57,880
So this tool would enable
to express oneself

63
00:03:58,200 --> 00:04:01,120
without giving one's identity,

64
00:04:01,440 --> 00:04:04,520
either from sign language,

65
00:04:04,840 --> 00:04:07,240
or from a written text.

66
00:04:07,560 --> 00:04:10,600
And then we'd be equal
to hearing people

67
00:04:10,920 --> 00:04:12,680
in terms of anonymity.

68
00:04:13,000 --> 00:04:14,040
What would you advise

69
00:04:14,360 --> 00:04:17,720
to make this project successful
and the software usable one day?

70
00:04:20,000 --> 00:04:22,680
So in order for the project
to succeed,

71
00:04:23,000 --> 00:04:27,320
I would advise in the first place
to work with the deaf world.

72
00:04:27,640 --> 00:04:30,640
Some hearing people know the deaf world,

73
00:04:30,960 --> 00:04:33,560
the sign language, its linguistics,

74
00:04:33,880 --> 00:04:37,800
but it seems important to me
to get the deaf involved

75
00:04:38,120 --> 00:04:42,040
and assess the results
of this tool with deaf people,

76
00:04:42,360 --> 00:04:45,160
because they will be the first users.

77
00:04:45,480 --> 00:04:49,440
So for me, this is the priority:
working with deaf people

78
00:04:49,760 --> 00:04:53,000
all along the project,
all the way to its end.

79
00:04:53,320 --> 00:04:56,920
It can only serve the project.
SYSTRAN