DISCLAIMER: this is a *sanity test* only, and is *not* to be regarded as a 
valid test of the VoxForge Acoustic Models!
 * the speech files used for testing use a single person's voice;
 * the same person also contributed speech for the creation of the VoxForge the 
   Acoustic Models;
 * there are only 50 audio samples in the test database - not enough for a 
   good test;
 * the VoxForge Acoustic Models are still an alpha release with respect to 
   Speaker Independent Speech Recognition, so please donate some speech to VoxForge.

Testing Acoustic Models created in: /data/svn-mirror/Nightly_Builds/AcousticModel-2021-01-04

HTK 8kHz_16bit
---------------
  Parameters:
	word insertion penalty: 0.0
	grammar scale factor: 1.0
====================== Results Analysis =======================
  Date: Mon Jan  4 02:32:53 2021
  Ref : /data/temp/_mionX1TIL/testref.mlf
  Rec : /data/temp/_mionX1TIL/recout.mlf
------------------------ Overall Results --------------------------
SENT: %Correct=68.00 [H=34, S=16, N=50]
WORD: %Corr=95.77, Acc=83.60 [H=181, D=1, S=7, I=23, N=189]
===================================================================

Julius 8kHz_16bit
------------------
  Parameters:
	word insertion penalty
	  first pass (-penalty1):0.5
	  second pass (-penalty2):100.0
	transition penalty (-iwsppenalty):-55.0 (for short-term inter-word pauses between words)
====================== Results Analysis =======================
  Date: Mon Jan  4 02:32:54 2021
  Ref : /data/temp/_mionX1TIL/testref.mlf
  Rec : /data/temp/_mionX1TIL/juliusProcessed
------------------------ Overall Results --------------------------
SENT: %Correct=70.00 [H=35, S=15, N=50]
WORD: %Corr=86.24, Acc=86.24 [H=163, D=17, S=9, I=0, N=189]
===================================================================



================================================================================================================
For comparison purposes, see below for the same Tests on the most current
release of the most current release of the VoxForge Acoustic Models:  
(/data/svn-mirror/Main/Tags/Releases//0.9.0)
================================================================================================================


HTK 8kHz_16bit
---------------
  Parameters:
	word insertion penalty: 0.0
	grammar scale factor: 1.0
====================== Results Analysis =======================
  Date: Mon Jan  4 02:32:59 2021
  Ref : /data/temp/_mionX1TIL/testref.mlf
  Rec : /data/temp/_mionX1TIL/recout.mlf
------------------------ Overall Results --------------------------
SENT: %Correct=68.00 [H=34, S=16, N=50]
WORD: %Corr=96.30, Acc=84.13 [H=182, D=1, S=6, I=23, N=189]
===================================================================

Julius 8kHz_16bit
------------------
  Parameters:
	word insertion penalty
	  first pass (-penalty1):0.5
	  second pass (-penalty2):100.0
	transition penalty (-iwsppenalty):-55.0 (for short-term inter-word pauses between words)
====================== Results Analysis =======================
  Date: Mon Jan  4 02:33:00 2021
  Ref : /data/temp/_mionX1TIL/testref.mlf
  Rec : /data/temp/_mionX1TIL/juliusProcessed
------------------------ Overall Results --------------------------
SENT: %Correct=72.00 [H=36, S=14, N=50]
WORD: %Corr=87.30, Acc=87.30 [H=165, D=16, S=8, I=0, N=189]
===================================================================

Notes:

    * the line starting with SENT gives the percentage of sentences that were 
      recognized correctly, out of N sentences in total.
    * the line starting with WORD gives the percentage of words that were 
      recognized correctly, out of N words in total
      However, since HTK or Julius erroneously 'added' words that are not in 
      the audio file (i.e. insertion errors) they usually get a lower 
      percentage accuracy rating.
    * Count definitions:
          o D - Deletion Error
          o S - Substitution Error
          o I - Insertion Error
