=====================================================
West Point Russian Speech 
=====================================================

Developers:   Col. Stephen A. LaRocca, Dr. Christine Tomei, John Morgan
Authors:  Col. Stephen A. LaRocca and Dr. Christine Tomei

The Center For Technology Enhanced Language Learning
Department Of Foreign Languages
United States Military Academy
745 Brewerton Road
West Point, NY 10996
Email: gs0416@usma.edu
Phone: 845-938-5286
Fax: 845-938-3585

Introduction:

Staff and Faculty of the Department of Foreign Languages (DFL) and the Center for Technology Enhanced Language Learning (CTELL) designed the West Point Russian corpus to provide a set of recordings for the training and development of speaker-independent speech recognition systems for use by West Point cadets enrolled in the Russian language program.

Materials:

The Collection Script:

The collection script consists of 96 sentences with a total of 528 tokens and 351 types. The prompts are labeled "s01" through "s96".  
	
The Lexicon:

The file "dict.txt" contains 690 distinct orthographic word forms, including all words found in the collection script. Each line of the lexicon contains one word entry: the orthography is given first, followed by a tab character, then the phone string for the word, with space characters separating the individual phone symbols.  
All phone strings end with the "sp" (short pause) segment.

The Phones:

@back mid , unrounded vowel   
A back low tense  vowel 
B voiced bilabial palatalized stop  
CH unvoiced palato--alveolar affricate 
D voiced dental palatalized stop 
E front mid  tense vowel 
F unvoiced labio--dental palatalized fricative 
G voiced  palatal stop 
I front high  tense vowel 
M bilabial palatalized nasal 
N dental  palatalized nasal  
O mid-back Round  vowel 
P unvoiced bilabial palatalized stop 
R alveolar palatalized discontinuous affricate 
S unvoiced dental palatalized fricative 
SH unvoiced palato--alveolar palatized fricative 
T unvoiced dental palatalized stop 
V voiced labio--dental palatized fricative 
Z voiced dental palatalized fricative 
a back lax  vowel      
b voiced bilabial stop 
c unvoiced dental affricate 
d voiced dental stop 
e mid--front lax vowel 
f unvoiced labio--dental fricative 
g voiced velar stop 
i front high  lax  vowel 
j semi--vowel consonantal glide 
k unvoiced velar stop 
l alveolar--dental liquid  
L alveolar--dental palatalized liquid  
m bilabial nasal  
n dental nasal 
p unvoiced bilabial stop 
r alveolar discontinuous affricate  
s unvoiced dental fricative 
sh unvoiced palato--alveolar  fricative 
sil silence 
sp short pause 
t unvoiced dental stop 
u back high  lax round  vowel 
v voiced labio--dental fricative 
x velar fricative      
y back high  lax unrounded   vowel 
z voiced dental fricative 
zh voiced palato--alveolar fricative 


The Transcriptions

Each waveform file has a monophone and word level master label file (*.mlf) transcription in HTK-format; the *.mlf files are presented in a concatenated form in "monophone.mlf" in the labels subdirectory. These files contain a multi-line entry for every speech file in the corpus -- the first line of each entry gives the file name, and the phones are provided in sequence on the following lines, one phone per line. Master label files are provided at both the word level and the phone level. All sentence transcripts begin and end with the "sil" (silence) segment. 

The Data:

Speech data  was collected using laptop computers running Windows NT.  
Recordings were captured at a sampling rate of 16 bit @ 22050 Hz using a Shure SM10A microphone and a RANE Model MS1 pre-amplifier.  A visual display  of the sentence, along with a digital recording of the sentence as read by a native speaker was presented. The informant pressed the Enter key to record their utterance. The informant's recording was played back for review, and the utterance was re-recorded, if necessary.  

The corpus consists of 4,181 speech files. Approximately 2,290 are from native informants and 1,891 files are from non-native informants. The following tables show the breakdown of corpus content in terms of male, female, native and non-native speakers.

     Table 1: number of speakers

		male   female   total
native:	13       16       29
non-native:	16       10       26
totals:	29       26       55


     Table 2: hours of data

		male   female   total
native:	1.1     1.3     2.4
non-native	1.1    0.8     1.9
totals:	2.2    2.1    4.3


     Table 3: Bytes of data

		male   		female    		total
native:	177932000		211848000		389776000
non-native:	181528000		134068000		315592000
totals	359456000		345908000		705364000


     Table 4: number of speech files

		male		female	total
native:	1027		1263		2290
non-native:	1103		788		1891
totals:	2130		2051		4181


Acknowledgments

Planning, execution and development of the West Point Russian Speech corpus was performed by the following members of the Center for Technology Enhanced Language Learning: John J Morgan, Col. Stephen A LaRocca, Charles Ruscelli, and Sherri Bellinger.


The following members of the Department Of Foreign Languages at West Point are acknowledged: Dr. Lawrence Mansour, Maj. David Bennett.

John Morgan thanks Dr. T.V. Raman for his audio interface to emacs called emacspeak.