Newsgroups: comp.speech
Path: pavo.csi.cam.ac.uk!pipex!uunet!elroy.jpl.nasa.gov!decwrl!deccrl!news.crl.dec.com!dbased.nuo.dec.com!nntpd.lkg.dec.com!nntpd2.cxo.dec.com!empros.cxn.dec.com!jimw
From: jimw@empros.cxn.dec.com (Jim West)
Subject: Re: Dialogic .VOX file format info wanted
Message-ID: <1993Jan4.195524.13376@nntpd2.cxo.dec.com>
Lines: 26
Sender: usenet@nntpd2.cxo.dec.com (USENET News System)
Reply-To: j_west@zebpik.enet.dec.com
Organization: Random Mixed
References:  <C06Kvo.E3E@news.cso.uiuc.edu> <1993Jan4.185746.11930@nntpd2.cxo.dec.com>
Date: Mon, 4 Jan 1993 19:55:24 GMT



  ...more specifically the files are in VBASE/40 format. (at least the
indexed type are)

  There is a 24 byte header and a 12 byte record for each prompt (may be
only one).  All data is stored as long integers (4 bytes each).

  Header Record:
	Field	Description
	  1	Maximum number of directory records
	  2	Sampling rate
	  3	Number of prompts used in the file
	  4	Reserved
	  5	Bytes used in file
	  6	Reserved

  Prompt Record:
	Field	Description
	  1	Offset in file to start of speech data
	  2	Length of prompt in bytes
	  3	Offset in file to start on annotation text

  Hope this helps

-/ Jim /-
