Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.java.programmer > #31342 > unrolled thread

For Java programmers looking for a cool project...

Back to article view | Back to comp.lang.java.programmer

Contents

  For Java programmers looking for a cool project... Ramon F Herrera <ramon@patriot.net> - 2015-03-20 19:53 -0500
    Re: Among other things, this is about Dilbert's Pointy-Haired Boss Ramon F Herrera <ramon@patriot.net> - 2015-03-20 19:58 -0500
    Re: For Java programmers looking for a cool project... Joerg Meier <joergmmeier@arcor.de> - 2015-03-21 10:17 +0100
      Re: For Java programmers looking for a cool project... Ramon F Herrera <ramon@conexus.net> - 2015-03-22 12:54 -0500
        Re: For Java programmers looking for a cool project... Ramon F Herrera <ramon@conexus.net> - 2015-03-22 14:10 -0500
      Re: For Java programmers looking for a cool project... John <xsli2@yahoo.com> - 2015-03-23 06:46 -0700
        Re: For Java programmers looking for a cool project... Joerg Meier <joergmmeier@arcor.de> - 2015-03-23 20:34 +0100
        Re: For Java programmers looking for a cool project... Ramon F Herrera <ramon@conexus.net> - 2015-03-23 18:48 -0500

#31342 — For Java programmers looking for a cool project...

[Note: this projects touches the realm of US politics - If not 
interested please proceed to the next post]

If you are somewhat familiar with CMU Sphinx, or would like to learn 
about it, this is a good opportunity.

 
https://sourceforge.net/p/cmusphinx/discussion/speech-recognition/thread/1d197b96/

TIA,

-Ramon

[toc] | [next] | [standalone]

#31343 — Re: Among other things, this is about Dilbert's Pointy-Haired Boss

On 3/20/2015 7:53 PM, Ramon F Herrera wrote:
>
> [Note: this projects touches the realm of US politics - If not
> interested please proceed to the next post]
>
> If you are somewhat familiar with CMU Sphinx, or would like to learn
> about it, this is a good opportunity.
>
>
> https://sourceforge.net/p/cmusphinx/discussion/speech-recognition/thread/1d197b96/
>
>
> TIA,
>
> -Ramon
>

See Dilbert's Boss:

http://goo.gl/xUrw2p

You may call this project "The Revenge of the Geeks".

-Ramon

[toc] | [prev] | [next] | [standalone]

#31344

On Fri, 20 Mar 2015 19:53:01 -0500, Ramon F Herrera wrote:

> [Note: this projects touches the realm of US politics - If not 
> interested please proceed to the next post]

> If you are somewhat familiar with CMU Sphinx, or would like to learn 
> about it, this is a good opportunity.

> https://sourceforge.net/p/cmusphinx/discussion/speech-recognition/thread/1d197b96/

That seems more like a project for a university grant, as this seems almost
purely research, and no actual programming. Or are there established
algorithms to recognize voices that only need implementing ?

Liebe Gruesse,
		Joerg

-- 
Ich lese meine Emails nicht, replies to Email bleiben also leider
ungelesen.

[toc] | [prev] | [next] | [standalone]

#31349

[Multipart message — attachments visible in raw view] — view raw

On 3/21/2015 4:17 AM, Joerg Meier wrote:
> On Fri, 20 Mar 2015 19:53:01 -0500, Ramon F Herrera wrote:
>
>> [Note: this projects touches the realm of US politics - If not
>> interested please proceed to the next post]
>
>> If you are somewhat familiar with CMU Sphinx, or would like to learn
>> about it, this is a good opportunity.
>
>> https://sourceforge.net/p/cmusphinx/discussion/speech-recognition/thread/1d197b96/
>
> That seems more like a project for a university grant, as this seems almost
> purely research, and no actual programming. Or are there established
> algorithms to recognize voices that only need implementing ?
>
> Liebe Gruesse,
> 		Joerg
>

There are several approaches to Speaker Identification.

This is a state of the art:

https://www.youtube.com/watch?v=5EHnNYwgo-U

This company charges $3500 for the job:

   http://www.owlinvestigations.com/index.html#

I figure that Sphinx has, somewhere inside its guts, speaker identification.

-Ramon

[toc] | [prev] | [next] | [standalone]

#31350

[Multipart message — attachments visible in raw view] — view raw

On 3/22/2015 12:54 PM, Ramon F Herrera wrote:
> On 3/21/2015 4:17 AM, Joerg Meier wrote:
>> On Fri, 20 Mar 2015 19:53:01 -0500, Ramon F Herrera wrote:
>>
>>> [Note: this projects touches the realm of US politics - If not
>>> interested please proceed to the next post]
>>
>>> If you are somewhat familiar with CMU Sphinx, or would like to learn
>>> about it, this is a good opportunity.
>>
>>> https://sourceforge.net/p/cmusphinx/discussion/speech-recognition/thread/1d197b96/
>>>
>>
>> That seems more like a project for a university grant, as this seems
>> almost
>> purely research, and no actual programming. Or are there established
>> algorithms to recognize voices that only need implementing ?
>>
>> Liebe Gruesse,
>>         Joerg
>>
>
> There are several approaches to Speaker Identification.
>
> This is a state of the art:
>
> https://www.youtube.com/watch?v=5EHnNYwgo-U
>
> This company charges $3500 for the job:
>
>    http://www.owlinvestigations.com/index.html#
>

Joerg:

If money were not an object, and I weren't interested in this as a great 
opportunity to gain a meaningful learning experience, what I would need 
to do with the O'Reilly audioclips is the first two types of analyzes, 
for a total of $8,000.

  • The first service is "Voice Identification", to ascertain whether 
the voice in the CNN clip indeed was produced by that gentleman's vocal 
chords and guttural tract: Owen charges $3,500 for this.

  • Additionally, the tape would be analyzed for "Authenticity": no 
editing, cuts, etc. Nixon kind of stuff. Owen charges $4,500 for this.

See attachment.

-Ramon

[toc] | [prev] | [next] | [standalone]

#31351

On Saturday, March 21, 2015 at 5:17:39 AM UTC-4, Joerg Meier wrote:
> On Fri, 20 Mar 2015 19:53:01 -0500, Ramon F Herrera wrote:
> 
> > [Note: this projects touches the realm of US politics - If not 
> > interested please proceed to the next post]
> 
> > If you are somewhat familiar with CMU Sphinx, or would like to learn 
> > about it, this is a good opportunity.
> 
> > https://sourceforge.net/p/cmusphinx/discussion/speech-recognition/thread/1d197b96/
> 
> That seems more like a project for a university grant, as this seems almost
> purely research, and no actual programming. Or are there established
> algorithms to recognize voices that only need implementing ?
> 
> Liebe Gruesse,
> 		Joerg
> 
> -- 
> Ich lese meine Emails nicht, replies to Email bleiben also leider
> ungelesen.

By googling "lie detection software open source", there are quite a few shows up.

The MIT one seems very cool:
http://www.extremetech.com/extreme/149623-mit-releases-open-source-software-that-reveals-invisible-motion-and-detail-in-video

[toc] | [prev] | [next] | [standalone]

#31354

On Mon, 23 Mar 2015 06:46:37 -0700 (PDT), John wrote:

> By googling "lie detection software open source", there are quite a few shows up.

From his linked thread:

>> A previous post of mine "From Speech Recognition to Lie Recognition?"
>> gave the incorrect impression that we are trying to analyze voice in
>> order to detect lies based in the signal content: i.e. some sort of
>> "Voice-based Lie Detector" contraption.

>> That is not it at all!

Liebe Gruesse,
		Joerg

-- 
Ich lese meine Emails nicht, replies to Email bleiben also leider
ungelesen.

[toc] | [prev] | [next] | [standalone]

#31358

On 3/23/2015 8:46 AM, John wrote:
> By googling "lie detection software open source", there are quite a few shows up.
>

Thanks, John.

As Joerg correctly pointed out, I am not looking for a lie detector.

This is what happened: in 1977 the subject told the truth, he was in 
Dallas and called his friend (a Congressional Investigator) in Florida:

"The Baron killed himself!!!??? Really? When? I gotta get a late flight 
to Florida tonight, or a car".

http://patriot.net/~ramon/jfk/MaddowOReilly.mp3 (audio only)
http://patriot.net/~ramon/jfk/MaddowOReilly.mp4 (video)

The actual lie occurred some 30 years later when he wrote a book claiming:

"When I was about to knock on his door, I heard a gunshot: the Russian 
Baron had committed suicide".

So, unless somebody comes up with a "text analyzing lie detector"  :-) ...

================================

Having said that, I followed your idea, Googled (for the 12th. time):

      "speaker identification software open source"

and this time I guess the query was in the proper order (or the cache 
changed?) because I made a *new* finding which looks very promising:

http://www.signalprocessingsociety.org/technical-committees/list/sl-tc/spl-nl/2013-05/ALIZE/

Thanks,

-Ramon

[toc] | [prev] | [standalone]

Back to top | Article view | comp.lang.java.programmer

csiph-web