I need to create a voice-controlled multimedia presentation, to be
presented in a kiosk. I am an experience multimedia developer, but
new to voice applications. I have read a bit on VoiceXML, which looks
perfect as far as having grammars to specify, leading to branching,
and NOT being user-dependent (no training). But this is for phone apps
only, and I haven't found a way to use it with a PC (or Mac) and a
presentation.
Question: what is a good set of software and hardware
to use to create and deliver this sort of application?
I don't use Macs,but I can if I have to, and I've heard they used to have
tools and examples in this area, but I can find nothing recent.
I have found lots of VoiceXML apps for phone applications,
but again nothing for driving displays and presentations.
(The solution can be simple and cheap, or even cost $$ up to some
reasonable level!) |
Request for Question Clarification by
jbf777-ga
on
08 May 2003 10:20 PDT
Hi -
Were you able to look at those other options at all?
|
Clarification of Question by
cpuworks-ga
on
08 May 2003 17:53 PDT
Not really, will try this weekend.
|
Request for Question Clarification by
pafalafa-ga
on
08 May 2003 19:33 PDT
Hello cpuworks,
I don't want to post this as an answer, because it is only one of
several suggestions you've received at this point. But from what I
can glean from your question, a very well-regarded piece of software
called NaturallySpeaking should do the trick, as long as you are in a
Windows-based system:
http://microURL.com/shrinkyTZIQ
"Dragon NaturallySpeaking® Solution Series speech recognition software
is a powerful desktop productivity tool that enables professional
users to work faster and delivers a compelling ROI to any
organization....Dragon NaturallySpeaking software works with virtually
all Windows®-based applications. Listen to incoming e-mails and
documents read aloud. Create your own documents on the spot or send
your recorded dictation to be transcribed elsewhere. Use on a PC or
with any ScanSoft™-certified handheld digital recorder....Speeds up
routine tasks on the PC by enabling you to create and format
documents, enter time and billing information, write and send emails,
and more—all by voice...Perform complex tasks with a spoken word or
phrase: Create custom voice commands (macros) that create documents,
insert boilerplate text, launch applications, fill in forms, and
automate complicated tasks with a single spoken word or phrase.
----------
Let us know which option seems the most promising (if any!!!), and we
can develop a full description of it as an answer to your question.
|
Request for Question Clarification by
pafalafa-ga
on
08 May 2003 19:46 PDT
Here's the link:
http://shop.voicerecognition.com/itemdesc.asp?CartId=808-ACCWARE-1308AVNYG122&ic=DNS+%2D+7%2E0+PROF&tpc=
My attempt to shrink it didn't work for some reason.
|
Request for Question Clarification by
pafalafa-ga
on
09 May 2003 06:19 PDT
Oh heck...just go here, and then click on the "NaturallySpeaking" box:
http://www.scansoft.com/naturallyspeaking/
|
Clarification of Question by
cpuworks-ga
on
09 May 2003 16:13 PDT
I have looked at the Naturally Speaking, but my first impulse is
that it requires training, and I want a system (like VoiceXML
in phone apps) that responds to more generic voices. The voice-enabled
PowerPoint suggestion from jbf777-ga may be the way to go, at least initially.
The presentation will be composed of a series of video clips (probably
mpeg), which should play one-after-the-other unless someone says one
of the magic words, like "more!" (when it will branch to a video with
more information on whatever the current topic is), or "again" (repeat
current segment), or "John Smith" (for example, if that's the person
being discussed in the mpeg), which would go to another mpeg with more
info on John Smith. The branching would be simple, with restricted cue words,
but different cue words and branching during each video. PowerPoint seems
to handle the videos OK, but I have never scripted for PowerPoint ala the
example page suggested, so I don't know what custom branching can be done.
But it looks good, and I think that's what I'm going to try for my first
attempt. There may be better solutions around, but I want to get ONE
working before I worry about getting a more optimal one. I will keep
you guys posted, and thanks for the pointers.
I guess figuring out which way works will determine who gets credit for
the answer!
|
Clarification of Question by
cpuworks-ga
on
10 May 2003 14:50 PDT
So far it looks like using the Microsoft Speech SDK with PowerPoint
might be the way to go, but I haven't installed the required software
yet, mainly because it's unclear where and how to get it.
Required OS: Windows 2000 SP3 - OK!
Required SW:
Microsoft Internet Information Services (IIS): can find the web page,
not sure how to get it and if it costs anything, and if so how much???
Microsoft Internet Explorer 6 or later: OK!
English MS Visual Studio .NET (various possible versions): $854 to buy
Microsoft .NET Framework Service Pack 2 (NOTE: NOT .NET Framework 1.1):
downloadable from Microsoft once other s/w obtained.
SpeechSDK itself: free massive d/l from Microsoft.
So that's at least one purchase, about $800, plus one big question (re IIS).
Has anyone out there used the SpeechSDK and gone through this process?
Pointers and experience appreciated.
|
Request for Question Clarification by
jbf777-ga
on
10 May 2003 18:37 PDT
Sound good! According to newsgroup threads, Microsoft IIS comes with
Windows.
Here are threads on it:
http://groups.google.com/groups?q=microsoft+internet+information+services&hl=en&lr=&ie=UTF-8&selm=u9zKIm%23nCHA.2388%40TK2MSFTNGP08&rnum=3
http://groups.google.com/groups?q=microsoft+internet+information+services&hl=en&lr=&ie=UTF-8&selm=014401c2fbce%24e788c450%24a301280a%40phx.gbl&rnum=6
These threads are both from this parent newsgroup discussion board:
microsoft.public.inetserver.iis
http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&group=microsoft.public.inetserver.iis
With regards to other users using Speech DSK, I would visit the
microsoft.public.speech_tech.sdk newsgroup discussion boards:
http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&group=microsoft.public.speech_tech.sdk
I would post any questions you have there, since there are aficionados
and quite possibly Microsoft MVP's who frequent that group.
If it's OK with you, I'll go ahead and post this as an answer?
jbf777-ga
|
Clarification of Question by
cpuworks-ga
on
12 May 2003 03:02 PDT
Sure, go ahead. It looks like there's no easy "install this one product
and you're ready to go", but one of the leads provided should do the trick.
|
