Blog Archive

Tuesday, March 31, 2015

order gawn

https://www.utdallas.edu/registrar/graduation/graduation-regalia/
http://www.delmonicohatter.com/sizing1.htm

Sunday, March 29, 2015

噪声库(noiseX-92)

噪声库(noiseX-92)

来自 Signal Processing Information Base (SPIB)
Noise data
Noise dataGo back to signal data directory selection.
Go back to data directory selection.
Go back to SPIB's main page.


2. NOIZEUS语音库:
来自于Loizou 的语音实验室http://www.utdallas.edu/~loizou/loires.html#recog
这个语音库有纯净的语音和不同信噪比的带噪语音,但是没有纯噪声。

3. TIMIT语音库:

4. 其他:
ETSI EG 202 396-1 V1.2.3 语音和多媒体传输质量(STQ);在背景噪音下的语音质量性能;第1部分:背景噪音模拟技术和背景噪音数据库
是一些双耳的噪声库,开放部分。

Friday, March 27, 2015

What would motivate you to try to be healthier?

 To live longer

 To have more energy or strength

 To feel better or happier

 To save money

"



'via Blog this'

Wednesday, March 25, 2015

Morgan Lira Obituary - San Jose, CA | San Jose Mercury News

Morgan Lira Obituary
Morgan Lira Obituary - San Jose, CA | San Jose Mercury News: "Morgan Lira
Resident of San Jose
Morgan Lira, 20, passed away Monday, March 16th at Louisville University Hospital, from injuries she suffered after being struck by a car while riding her bike.
Morgan was born in San Jose, California. She is a graduate of Valley Christian High School. Morgan moved to Kentucky to attend Morehead State University on an athletic scholarship.
She was a proud member of the Kentucky National Guard, which became her second family.
Morgan loved running, working out, skateboarding and surfing. Morgan was a warm and beautiful girl both inside and out. She touched many lives with her giving heart and beautiful smile in her short time here.
Morgan is survived by her parents, Rich and Lisa, her sister Olivia, brother Michael as well as a large extended family and many wonderful friends.
A memorial service will be held at Church on the Hill, Sunday evening March 29th at 6:30 p.m.
Published in San Jose Mercury News/San Mateo County Times on Mar. 25, 2015"



'via Blog this'

Monday, March 23, 2015

Tuesday, March 17, 2015

About two VAD toolkit

Harmfreq_MOLRT: a statistical model, likelihood ratio test (LRT)-based speech/non-speech detection algorithm
Harmfreq_MOLRT is a statistical model, likelihood ratio test (LRT)-based speech/non-speech detection algorithm. The likelihood ratios (LRs) for voiced and unvoiced frames are computed differently: LR for voiced frames is calculated using only the harmonic DFTs; for unvoiced frames, LR is calculated using all DFTs.  It is an improved version of the multiple observation (MO) LRT VAD proposed by Ramirez et. al. [Matlab code of Harmfreq_MOLRT VAD]
MBSC: a Multi-Band Summary Correlogram (MBSC)-based pitch detection algorithm for noisy speech
MBSC is a Multi-Band Summary Correlogram (MBSC)-based pitch detection algorithm for noisy speech.  The package contains the matlab code that is used to generate the pitch detection results reported in L. N. Tan, and A. Alwan, "Multi-Band Summary Correlogram-based Pitch Detection for Noisy Speech", Speech Communication, in press.  A fast version of the code is also provided in the package. [Matlab code of MBSC pitch detector]
http://www.seas.ucla.edu/spapl/shareware.htm

'via Blog this'

Wednesday, March 11, 2015

ALIZE ivector demo GMM UBM

Softwares:

Reference:
http://www1.i2r.a-star.edu.sg/~alarcher/Softwares.html

Tutorial on GMM / UBM for ALIZE 3.0 (download)
In this archive you will find everything required to create your own GMM/UBM system for speaker recognition using ALIZE toolkit. The aim of this demo is to provide an example how to combine ALIZE components in order to learn your Universal Background Model, adapt speaker dependent GMMs, test against speech utterances and normalize the scores.

Config files are just examples and have to be modified in order to produce state-of-the-art-performance. I would like to thank LDC and the NIST for allowing the use of NIST-SRE 2004 data in this tutorial. SRE04 was created by NIST from data and annotations provided by LDC. Sponsors and participants of the NIST Speaker Recognition Evaluations have also shaped the corpus through their requirements, analysis and feedback.


# The following spro feature extraction based GMM UBM framework:
date
bash 01_RUN_feature_extraction.sh > output1.txt
date
bash 02a_RUN_spro_front-end.sh > output2.txt
date
bash 03_RUN_gmm-ubm.sh > output3.txt
date



'via Blog this'

htk installation problem on ubuntu 14.04 - Ask Ubuntu

htk installation problem on ubuntu 14.04 - Ask Ubuntu:



Question:



I have already download and unpack htk-3.4.1, then according to the htk installation guide I change the directory to this file then type "./configure --prefix=/usr/local/bin.linux" command. it work fine and provide me two necessary commands to install the tool; these are: 1. make all to build htk and 2. make install to install htk, however when I enter the first command it displays this error
(cd HTKLib && make HTKLib.a) \ || case "" in k) fail=yes;; *) exit 1;; esac; make[1]: Entering directory /tmp/htk/HTKLib'
gcc -ansi -D_SVID_SOURCE -DOSS_AUDIO -D'ARCH="i686"' -Wall -Wno-switch -g -O2 -I. -DPHNALG -c -o HGraf.o HGraf.c
HGraf.c:73:77: fatal error: X11/Xlib.h: No such file or directory
#include <X11/Xlib.h> /* the X11 stuff makes string.h also available */
^
compilation terminated.
make[1]: *** [HGraf.o] Error 1
make[1]: Leaving directory
/tmp/htk/HTKLib' make: * [HTKLib/HTKLib.a] Error 1
please help me

Solution:

had the same problem installing HTK on Ubuntu 14.04. I found the answer on Voxforge forum. The setup was designed for 32-bit computers, and to install you need to call a 32 bash. Type this on a terminal window:
linux32 bash
Then proceed with the installation:
./configure --prefix=/home/yourname/htk
make all
sudo make install
You may also need to install 'libx11-dev', before typing these:
sudo apt-get install libx11-dev
http://askubuntu.com/questions/452208/htk-installation-problem-on-ubuntu-14-04
Note:
The final binary file can be find:
/home/yourname/htk/bin
Testing:

By type the following command,

/home/yourname/htk/bin/HCopy

You should see







USAGE: HCopy [options] src [ + src ...] tgt ...



 Option                                       Default



 -a i     Use level i labels                  1

 -e t     End copy at time t                  EOF

 -i mlf   Save labels to mlf s                null

 -l dir   Output target label files to dir    current

 -m t     Set margin of t around x/n segs     0

 -n i [j] Extract i'th [to j'th] label        off

 -s t     Start copy at time t                0

 -t n     Set trace line width to n           70

 -x s [n] Extract [n'th occ of] label  s      off

 -A      Print command line arguments         off

 -C cf   Set config file to cf                default

 -D      Display configuration variables      off

 -F fmt  Set source data format to fmt        as config

 -G fmt  Set source label format to fmt       as config

 -I mlf  Load master label file mlf

 -L dir  Set input label (or net) dir         current

 -O      Set target data format to fmt        as config

 -P      Set target label format to fmt       as config

 -S f    Set script file to f                 none

 -T N    Set trace flags to N                 0

 -V      Print version information            off

 -X ext  Set input label (or net) file ext    lab


Tuesday, March 10, 2015

The 3rd `CHiME' Speech Separation and Recognition Challenge

The 3rd `CHiME' Speech Separation and Recognition Challenge:



'via Blog this'

Perl Subroutines

Perl Subroutines:



A Perl subroutine or function is a group of statements that together perform a task. You can divide up your code into separate subroutines. How you divide up your code among different subroutines is up to you, but logically the division usually is so each function performs a specific task.
Perl uses the terms subroutine, method and function interchangeably.

Define & Call a Subroutine:

The general form of a subroutine definition in Perl programming language is as follows:
sub subroutine_name{
   body of the subroutine
}
The typical way of calling that Perl subroutine is as follows:
subroutine_name( list of arguments );
In versions of Perl before 5.0 the syntax for calling subroutines was slightly different as shown below. This still works in the newest versions of Perl, but it is not recommended since it bypasses subroutine prototypes.
&subroutine_name( list of arguments );
Let's have a look into the following example which defines a simple function and then call it. Because Perl compiles your program before executing it, it doesn't matter where you declare your subroutin.
#!/usr/bin/perl

# Function definition
sub Hello{
   print "Hello, World!\n";
}

# Function call
Hello();
When above program is executed, it produces following result:
Hello, World!

Passing Arguments to a Subroutine:

You can pass various arguments to a subroutine like you do in any other programming language and they can be acessed inside the function using the special array @_. Thus the first argument to the function is in $_[0], the second is in $_[1], and so on.
You can pass arrays and hashes as arguments like any scalar but passing more than one array or hash normally causes them to lose their separate identities. So we will use references ( explained in the next chapter ) to pass any array or hash.
Let's try following example which takes a list of numbers and then prints their average:
#!/usr/bin/perl

# Function definition
sub Average{
   # get total number of arguments passed.
   $n = scalar(@_);
   $sum = 0;

   foreach $item (@_){
      $sum += $item;
   }
   $average = $sum / $n;

   print "Average for the given numbers : $average\n";
}

# Function call
Average(10, 20, 30);
When above program is executed, it produces following result:
Average for the given numbers : 20

PASSING LISTS TO SUBROUTINES

Because the @_ variable is an array, it can be used to supply lists to a subroutine. However, because of the way in which Perl accepts and parses lists and arrays, it can be difficult to extract the individual elements from @_. If you have to pass a list along with other scalar arguments then make list as the last argument as shown below:
#!/usr/bin/perl

# Function definition
sub PrintList{
   my @list = @_;
   print "Given list is @list\n";
}
$a = 10;
@b = (1, 2, 3, 4);

# Function call with list parameter
PrintList($a, @b);
When above program is executed, it produces following result:
Given list is 10 1 2 3 4

PASSING HASHES TO SUBROUTINES

When you supply a hash to a subroutine or operator that accepts a list, then hash is automatically translated into a list of key/value pairs. For example:
#!/usr/bin/perl

# Function definition
sub PrintHash{
   my (%hash) = @_;

   foreach my $key ( keys %hash ){
      my $value = $hash{$key};
      print "$key : $value\n";
   }
}
%hash = ('name' => 'Tom', 'age' => 19);

# Function call with hash parameter
PrintHash(%hash);
When above program is executed, it produces following result:
name : Tom
age : 19

Returning Value from a Subroutine:

You can return a value from subroutine like you do in any other programming language. If you are not returning a value from a subroutine then whatever calculation is last performed in a subroutine is automatically also the return value.
You can return arrays and hashes from the subroutine like any scalar but returning more than one array or hash normally causes them to lose their separate identities. So we will use references ( explained in the next chapter ) to return any array or hash from a function.
Let's try following example which takes a list of numbers and then returns their average:
#!/usr/bin/perl

# Function definition
sub Average{
   # get total number of arguments passed.
   $n = scalar(@_);
   $sum = 0;

   foreach $item (@_){
      $sum += $item;
   }
   $average = $sum / $n;

   return $average;
}

# Function call
$num = Average(10, 20, 30);
print "Average for the given numbers : $num\n";
When above program is executed, it produces following result:
Average for the given numbers : 20

Private Variables in a Subroutine:

By default, all variables in Perl are global variables which means they can be accessed from anywhere in the program. But you can create private variables called lexical variables at any time with the myoperator.
The my operator confines a variable to a particular region of code in which it can be used and accessed. Outside that region, this variable can not be used or accessed. This region is called its scope. A lexical scope is usually a block of code with a set of braces around it, such as those defining the body of the subroutine or those marking the code blocks of if, while, for, foreach, and eval statements.
Following is an example showing you how to define a single or multiple private variables using myoperator:
sub somefunc {
   my $variable; # $variable is invisible outside somefunc()
   my ($another, @an_array, %a_hash); # declaring many variables at once
}
Let's check following example to distinguish between global and private variables:
#!/usr/bin/perl

# Global variable
$string = "Hello, World!";

# Function definition
sub PrintHello{
   # Private variable for PrintHello function
   my $string;
   $string = "Hello, Perl!";
   print "Inside the function $string\n";
}
# Function call
PrintHello();
print "Outside the function $string\n";
When above program is executed, it produces following result:
Inside the function Hello, Perl!
Outside the function Hello, World!

Temporary Values via local()

The local is mostly used when the current value of a variable must be visible to called subroutines. A local just gives temporary values to global (meaning package) variables. This is known as dynamic scoping. Lexical scoping is done with my, which works more like C's auto declarations.
If more than one variable or expression is given to local, they must be placed in parentheses. This operator works by saving the current values of those variables in its argument list on a hidden stack and restoring them upon exiting the block, subroutine, or eval.
Let's check following example to distinguish between global and local variables:
#!/usr/bin/perl

# Global variable
$string = "Hello, World!";

sub PrintHello{
   # Private variable for PrintHello function
   local $string;
   $string = "Hello, Perl!";
   PrintMe();
   print "Inside the function PrintHello $string\n";
}
sub PrintMe{
   print "Inside the function PrintMe $string\n";
}

# Function call
PrintHello();
print "Outside the function $string\n";
When above program is executed, it produces following result:
Inside the function PrintMe Hello, Perl!
Inside the function PrintHello Hello, Perl!
Outside the function Hello, World!

State Variables via state()

There another type of lexical variables which are similar to private variable but they maintain their state and they do not get reinitialized upon multiple calls of the subroutines. These variables are defined usingstate operator and available starting from Perl 5.9.4
Let's check following example to demonstrate the use of state variables:
#!/usr/bin/perl

use feature 'state';

sub PrintCount{
   state $count = 0; # initial value

   print "Value of counter is $count\n";
   $count++;
}

for (1..5){
   PrintCount();
}
When above program is executed, it produces following result:
Value of counter is 0
Value of counter is 1
Value of counter is 2
Value of counter is 3
Value of counter is 4
Prior to Perl 5.10 you would have to write it like this:
#!/usr/bin/perl

{
   my $count = 0; # initial value

   sub PrintCount {
      print "Value of counter is $count\n";
      $count++;
   }
}

for (1..5){
   PrintCount();
}

Subroutine Call Context

The context of a subroutine or statement is defined as the type of return value that is expected. This allows you to use a single function that returns different values based on what the user is expecting to receive. For example, the following localtime() returns a string when it is called in scalar context but it returns a list when it is called in list context.
my $datestring = localtime( time );
In this example, the value of $timestr is now a string made up of the current date and time, for example, Thu Nov 30 15:21:33 2000. Conversely:
($sec,$min,$hour,$mday,$mon, $year,$wday,$yday,$isdst) = localtime(time);
Now the individual variables contain the corresponding values returned by localtime() subroutine.

Wednesday, March 4, 2015

近期驾照问题汇总贴+跨州搬家换车牌驾照经验分享

http://www.moonbbs.com/thread-502096-1-1.html

以下提供一些最近经常出现的关于驾照问题的帖子,有相关问题,请浏览帖子下的回复。
先share一下我自己的经验。
(1) 到美国的第一学期考驾照,地点纽约,身份F1。先通过笔试,然后买车,预约路考,通过路考之后得到驾照。(考驾照时我还没有SSN,所以要先去SSA开一份证明。其他资料请上网google,有详细列表)


(2) 毕业后从NY搬家到CA。找运输公司运车,价格$1350, door to door service. 具体就是上网搜索找到好评多的公司,打电话咨询,预约NY取车时间,交给对方钥匙,签一份运输合同,预支300运输费。然后把不急需的行李放满整个后备箱(长途搬家运输衣服鞋子的最省钱的办法~呵呵)然后自己拖两个行李从NY飞到CA。5天后车到,检查车况(上面都是土。。。的确有点脏~)然后检查后备箱行李,没问题就支付剩余款,拿钥匙。

(3) 更换车牌和驾照。地点加州湾区,身份OPT。 加州换车牌需要预先做好SMOG test,拿到报告。然后上网预约DMV的appointment time,准备好所有你能想到的资料,包括你自己的和车的。(Tips:下载DMV的APP可以看到即时更新的排队等候时间~如果不想预约也可以用这个软件看一下什么时候人少,直接过去就好了。)
Mock test can be found via:
http://www.ccyp.com/TRAFFIC/

到了之后先进去领取表格,然后出门开车绕到DMV后面等待工作人员为车做检测。(具体的开过去就知道了,很简单,没什么可说的)然后对方会填好你手里的表格并签字。然后开车绕回DMV停车场,拿表格领取号码,等待叫号。叫到你的时候过去提交对方所需的所有材料,对方会当场给你CA的车牌。如果你不知道NY的车牌如何邮寄回去,就出门,当场把车牌卸下来,然后交给前台,请他们转交NY的DMV就可以了。


然后到DMV的另一边去拍照,领取笔试题目。(NY驾照换CA驾照,只需要考笔试。)笔试当场即可知道是否通过,通过了之后就交钱填表格留地址就好了。

在这里要特别说明一下申请驾照提交的材料和驾照有效期问题。
(1) 很多人问OPT期间是否需要EAD卡。根据我自己的经验,要的。不止是要,而且要等到EAD卡生效之后。所以,我到了加州一个月之后才被允许考驾照笔试。
(2) 看到很多人说自己的驾照过期了,EAD没到或者OPT失效,I20失效什么的无法更新驾照。
不知道你们当初考驾照的时候驾照的过期日期DMV的工作人员是如何决定的。我可能是个奇葩的个例,因为我2013年7月以OPT身份拿到CA驾照,驾照的有效期持续到2018年3月。过期日期是我的生日,与EAD或者OPT等有效期无关。所以期间不管什么证件过期了,应该都不会影响我的驾照过期日。可能的话,如果为了避免证件过期影响驾照,跟DMV的工作人员咨询一下,能否给与比较长的驾照有效期。

Monday, March 2, 2015

[matlab ubuntu installation] libc.so.6: not found



Question: 
I just installed Matlab R2010b on Ubuntu 14.04.
Upon initiation of Matlab, the following message is printed:
/usr/bin/matlab: 1: /usr/local/MATLAB/R2010b/bin/util/oscheck.sh: /lib64/libc.so.6: not found

Solution:

sudo ln -s /lib/x86_64-linux-gnu/libc-2.19.so /lib64/libc.so.6



Reference:



http://www.mathworks.com/matlabcentral/answers/43604-what-library-should-be-used-by-matlab-2012a-on-ubuntu-12-04-instead-of-libc-so-6


'via Blog this'