The '''Spoken Language ID''' project seeks to....

==Introductory Tasks==
===Running An Experiment in Ten Shell Commands Or Less===
# Ensure that your system meets the current [[Requirements | system requirements]]. On an Ubuntu Linux system, this can be done using a command such as this:<br/><code>sudo apt-get install sun-java6-jdk ant perl ruby gnuplot gcc cmake make pdl subversion</code>
# Check out the current Language ID repository:<br/><code>svn co <nowiki>http://nlp.cs.byu.edu/subversion/NIST/HEAD</nowiki></code>
# Enter the HEAD directory created by the previous command:<br/><code>cd HEAD</code>
# Be sure that the template from which the main configuration file will be generated represents your setup by editing <code>Language-ID/config/language-id.conf.cmake</code>. Do this using your favorite text editor, for example: <code>vim Language-ID/config/language-id.conf.cmake</code>. Pay particular attention to <code>PRAAT_EXE</code>, <code>PERL_EXE</code>, <code>WAV_DATA_ORIGINAL_LOCATION</code>, and <code>SEG_DATA_ORIGINAL_LOCATION</code> (or their <code>*_WIN</code> counterparts if running on Windows) to be sure that these contain the correct paths. (<code>LABTOOLS</code> seems to be unused at the moment.)
# Generate the build system with default settings using cmake:<br/><code>cmake .</code><br/>This also generates the configuration file <code>Language-ID/config/language-id.conf</code> based on the settings you provided in <code>Language-ID/config/language-id.conf.cmake</code>.
# Build DETware:<br/><code>make</code>
# Build Language-ID:<br/><code>cd Language-ID<br/>ant</code> 
# Run the experiment:<br/><code>cd ..<br/>make detcurve</code><br/>The <code>detcurve</code> target will create the <code>experiments</code> directory in which all experiment data will be stored. It will then copy seg files and wav files into <code>experiments/data</code> and begin extracting features from these data files by running the [[seg2xml3.pl]] script.<br/>Once these preliminary data have been copied and analyzed, a directory specific to the current experiment will be created. As the default experiment is <code>fourgramall</code>, a directory called <code>experiments/fourgramall</code> will be created to house results specific to that experiment. ling files will be generated and placed in <code>experiments/fourgramall/data</code>. Language models will be trained and placed in <code>experiments/fourgramall/models</code>. Other results, including metrics and plot data, will be placed in <code>experiments/fougramall/results</code>
# Explore the DET curves, avgcost.txt, avgeer.txt, etc. in the <code>results</code> directory.

===Running a Specific Experiment===
The above example only runs the default 'fourgramall' experiment. Running a specific experiment requires almost exactly the same process. For example, to run the 'fivegram' experiment, substitute the following cmake command:
:<code>cmake -D FEATURE_SET_NAME_FORCE=fivegram .</code>

Then proceed to build the <code>detcurve</code> target as before:
:<code>make detcurve</code>

This time, results will be stored in <code>experiments/fivegram</code> rather than <code>experiments/fourgramall</code> as previously.

Other parameters can be set, such as the normalization option for [[resultbuilder.pl]], and the result name that determines where plot output is stored:
:<code>cmake -D FEATURE_SET_NAME_FORCE=fivegram -D RBLDR_NORM_FORCE=1 -D RESULT_NAME_FORCE=nist_norm1 .</code><br/>
This prepares the system for running <code>fivegram</code> with a normalization of 1. It will also set the result name to be <code>nist_norm1</code> to differentiate this run from our previous one. We can now run the experiment:
:<code>make detcurve</code>

The parameters we set in our most recent run of cmake are the exact parameters used by the regression test. Speaking of the regression test.... <segue>

===Running All Experiments===
If you wish to build all defined experiments, simply run <code>Language-ID/scripts/[[runall.rb]]</code>.

===Regression Tests===
Regression tests have been created to help verify the integrity of any changes we make to the system. The regression tests can be run by invoking <code>Language-ID/scripts/[[run_all_tests.rb]]</code> from within the <code>HEAD</code> directory. This will attempt to compare all currently-built experiments to the baseline data stored in <code>Language-ID/regression</code> to guarantee no significant changes to the output.

==Papers==
The following papers provide vital background information to the problem that the Spoken Language ID project is tackling:
* [http://ieeexplore.ieee.org.erl.lib.byu.edu/iel4/89/10293/00481450.pdf?isnumber=&arnumber=481450 Zissman '96: Comparison of Four Approaches to Automatic Language Identification of Telephone Speech]
* [http://www.nist.gov/speech/tests/lre/2007/LRE07EvalPlan-v8b.pdf NIST evaluation guidelines]
* BYU NLP Lab [http://nlp.cs.byu.edu/techreports/BYUNLP-TR1.pdf Language ID Tech Report]

[[Category:Spoken Language ID]]