FreeTTS-1.2.2

Introduction to FreeTTS

The FreeTTS package contains a speech synthesis system written entirely in the Java programming language. It is based upon Flite: a small run-time speech synthesis engine developed at Carnegie Mellon University. Flite is derived from the Festival Speech Synthesis System from the University of Edinburgh and the FestVox project from Carnegie Mellon University. The FreeTTS package is used to convert text to audible speech through the system audio hardware.

This package is known to build and work properly using an LFS-7.4 platform.

Package Information

Additional Downloads

FreeTTS Dependencies

Required

apache-ant-1.9.2 and Sharutils-4.13.5

User Notes: http://wiki.linuxfromscratch.org/blfs/wiki/freetts

Installation of FreeTTS

The FreeTTS package is distributed in ZIP format and the unzip command will default to creating an unused source directory. Additionally, unzipping the test suite file will prompt for questions about overwriting existing files. Use the following commands to unzip the source files:

unzip -q freetts-1.2.2-src.zip -x META-INF/* &&
unzip -q freetts-1.2.2-tst.zip -x META-INF/*
[Tip]

Tip

The sh jsapi.sh command below installs the Java Speech API components into the FreeTTS source tree. You will be required to view, and then accept (by entering a y keypress), a license agreement before the installation will continue. If you are scripting (automating) the build, you'll need to account for this. There is information about automating build commands in the Automated Building Procedures section of Chapter 2. Towards the end of this section, specific information for automating this type of installation is discussed.

Install FreeTTS by running the following commands:

sed -i 's/value="src/value="./' build.xml &&
cd lib      &&
sh jsapi.sh &&
cd ..       &&
ant

To test the results, issue:

ant junit &&
cd tests &&
sh regression.sh &&
cd ..

Now, as the root user:


install -v -m755 -d /opt/freetts-1.2.2/{lib,docs/{audio,images}} &&
install -v -m644 lib/*.jar /opt/freetts-1.2.2/lib                &&
install -v -m644 *.txt RELEASE_NOTES docs/*.{pdf,html,txt,sx{w,d}} \
                               /opt/freetts-1.2.2/docs           &&
install -v -m644 docs/audio/*  /opt/freetts-1.2.2/docs/audio     &&
install -v -m644 docs/images/* /opt/freetts-1.2.2/docs/images    &&
cp -v -R javadoc               /opt/freetts-1.2.2                &&
ln -v -s freetts-1.2.2 /opt/freetts

Optionally, install any or all of the additional FreeTTS components using the following commands as the root user (see the Command Explanations section for details):

cp -v -R bin    /opt/freetts-1.2.2        &&
install -v -m644 speech.properties $JAVA_HOME/jre/lib &&
cp -v -R tools  /opt/freetts-1.2.2        &&
cp -v -R mbrola /opt/freetts-1.2.2        &&
cp -v -R demo   /opt/freetts-1.2.2

Command Explanations

sed -i 's/value="src/value="./' build.xml: Fix an error in the build file to allow the program to find the source.

sh jsapi.sh: This command installs the Java Speech API components into the FreeTTS source tree.

ant: FreeTTS uses the Apache Ant build system instead of the GNU autotools. This commands builds everything, including the class libraries, tools and demos.

cp -v -R bin ...; install -v -m644 speech.properties: These two commands install the demonstration programs. Optionally copy the speech.properties file to ~/speech.properties if you don't want to make it available system-wide.

cp -v -R tools ...: This installs the voice data import utilities. See the README.html files in the tools/ subdirectories for information and instructions about using the tools.

cp -v -R mbrola ...: This installs the mbrola.jar file, required if you use the MBROLA voices.

cp -v -R demo ...: This installs the sources and documentation for the demonstration programs.

For additional information and documentation about the FreeTTS project, visit the main web page at http://freetts.sourceforge.net.

Testing the Installation

Test the installation using the following command:

java -jar /opt/freetts/lib/freetts.jar \
    -text "This is a test of the FreeTTS speech synthesis system"

Depending on the setup of your audio drivers and software, you may have to add the -streaming switch to the command as shown below:

java -jar /opt/freetts/lib/freetts.jar -streaming \
    -text "This is a test of the FreeTTS speech synthesis system"

Contents

Installed Programs: None
Installed Libraries: /opt/freetts-1.2.2/lib/*.jar
Installed Directory: /opt/freetts-1.2.2

Short Descriptions

*.jar

contains the class libraries which make up the FreeTTS speech synthesis system.

Last updated on 2013-09-01 15:01:22 -0700