BAS

                   Bavarian Archive for Speech Signals

               University of Munich, Institut of Phonetics
               Schellingstr. 3/II, 80799 Munich, Germany
                    bas@phonetik.uni-muenchen.de


         COPYRIGHT University of Munich 1995. All rights reserved.   
This corpus and software may not be disseminated further - not even
partly - without a written permission of the copyright holders.  

                    Additional Copyright Holders

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

PhonDat Data Format  -  Description

F. Schiel	03.03.95 / 11.01.96

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

FOR THE FAST READER:

Use the ANSI C function in softw/header.c and data.c to read PhonDat
files. 

PHONDAT FILE FORMAT

Speech files archived in the PhonDat format have the following structure:

PhonDat Format 1

Bytes	Contents
1-512	Fixed length binary header structure, compatible with ILS format
513-EOF	nspbk blocks of 512 bytes, containing 256 short samples each, 
	Intel format.

PhonDat Format 2

Bytes	Contents
1-512	Fixed length binary header structure, compatible with ILS format
513-X	multiple of 512 Bytes, containing the orthography and the so-called
	canonical form of the utterance.
	Struture:
	512 - 515 			: o r t \0
  	516 - 516+no			: (no bytes orthography)
	516+no+1 - 516+no+12		: \0 o e n d : : \0 k a n \0
	516+no+13 - 516+no+13+nc	: (nc bytes canonical form (SAM-PA)
	516+no+13+nc+1 - 516+no+13+nc+8 : \0 k e n d : : \0
	516+no+13+nc+9 - X		: filled up with \0
X+1-EOF	nspbk blocks of 512 bytes, containing 256 short samples each,
        Intel format.

Fixed length part of PhonDat header (version 1 and 2)

Contains information about the filestructure, the data type, 
the recording conditions and the speaker. 
Use functions from softw/header.c for reading or
writing of PhonDat headers. 

The structure is as follows (C notation):

typedef struct {
                long    not_used1[5],
                        nspbk,          /* # of data blocks (512 bytes) */
                        anz_header,     /* # of header blocks (512 bytes) */
                        not_used2[5];
                char    sprk[2];        /* speaker id                   */
                short   swdh;           /* session repetition           */
                long    ifld1[3],       /* ILS                          */
                        not_used3[6];
                char    kenn1[2];       /* ILS text characters 1 - 8    */
                short   not_used4;
                char    kenn2[2];
                short   not_used5;
                char    kenn3[2];
                short   not_used6;
                char    kenn4[2];
                short   not_used7;
                long    not_used8[35],
                        isf,            /* sampling rate in Hz          */
                        flagtype,       /* ILS: -32000 if sampling file
                                                -29000 if param. file   */
                        flaginit;       /* ILS: 32149 if init           */
                char    ifl[32],        /* filename (terminated by /0)  */
                        day,            /* # of day                     */
                        month;          /* # of month                   */
                short   year;           /* # of year                    */
                char    sex,            /* sex of speaker               */
                        version;        /* header version:
                                           0 = extended ILS
                                           1 = phondat version 1
                                           2 = phondat version 2        */
                short   adc_bits,       /* resolution of adc            */
                        words;          /* # of words in text           */
                long    not_useda[50];
                short   wdh,            /* # of repetition              */
                        abs_ampl;       /* maximum of amplitude         */
                char    not_usedb[10];
                } Phon_header_2;

Variable lenght part of PhonDat header (version 2 only)

The orthographic string contains the standard orthography or a
transliteration with additional markers of the spoken utterance. 
German Umlauts are represented either by LaTeX 
convention or by 7 bit ASCII signs or by German Character set 
coding used by DEC and Sun:

Umlaut	LaTeX	7 Bit ASCII (dec) 	German Char Set (hex)
Ae	"A	[ (91)			C4
Ue	"U	] (93)			CD
Oe	"O	\ (92)			D6
ae	"a	{ (123)			E4
ue	"u	} (125)			FC
oe	"o	| (124)			F6
ss	"s	~ (126)			DF

The canonical string contains the exspected citation forms of the
word in the utterance. Note that this is NOT a transcription of the
signal. Symbols used are the German subcorpus of the SAM-PA, with 
following changes to SAM-PA:

Q	Glottal Stop
q	Glottalisierung (not in canonical forms!)
'	main stress
"	secondary stress
#	compositum marker (optional)
+ 	function word marker (suffix, optional)

Words are seperated by two blanks, phonemic labels are seperated by
one blank.