
The eSpeak homepage is:  http://espeak.sourceforge.net


Compiling
=========

The  src  directory contains a simple  makefile  that compiles all the
source files in the directory and produces a  "speak"  binary.

When run, the "speak" program expects to find the  espeak-data  directory
either in the user's home directory, or if not there, in /use/share.


Library Version.
================

The makefile also compiles a shared library version of eSpeak and creates a
file:  libespeak.so.1.xx  where xx is the version number.  This is not
currently used by the "speak" program, but is available for use by other
software (eg. the next version of Speech Dispatcher).

The API is described in the file:  speak_lib.h.


Portaudio.
========== 

The "speak" program uses the PortAudio sound interface library.  There are two
versions, V18 and V19 and their APIs are different.  Some Linux distributions
use V18 (eg. Debian, Ubuntu) others (eg. SuSe) use V19.

To compile to use a particular version of the PortAudio library, first copy
either  portaudio18.h  or  portaudio19.h  to  portaudio.h.

If you don't need eSpeak to output sound, but only to produce WAV files,
then you can remove or comment out the following line in  speech.h.
This will mean that eSpeak doesn't use any PortAudio functions.
 #define  USE_PORTAUDIO


Compiling Data
==============

You can modifiy spelling-to-phoneme rules and exceptions and re-compile this
data with the  speak  program (see  docs/dictionary.html).

In order to modify the sounds or other characteristics of phonemes, or
to add additional phonemes, another program, espeakedit, is needed, which
is not part of this package.

