README

This is a release of Dhvani, a Text-to-Speech software for 
Indian languages ( version 0.6 )

This text-2-speech distribution for Indian Languages contains

1. A phoneme to speech engine in the simputer/ subdirectory.
Running ./direct-synth and giving phonetic input on the prompt
will produce speech. The phonetic input format is described
in doc/README. Examples of phonetic texts are in the pdemos/
subdirectory.

2. A Unicode-Text-Format (UTF) to phonetics convertor for 
Hindi, called simputer/hindiphonserv&. This is designed to be a server
which will accept utf text from a client, convert in to
phonetics, and pass it on to the speech engine server, ./synthserv&.
An example of a UTF hindi text is in tdemos/hindi/utf.
An example of a client which will sent UTF to  hindiphonserv
is in simputer/client. See README in simputer/ to find out
how to invoke this client.

3. A Unicode-Text-Format (UTF) to phonetics convertor for 
Kannada, called simputer/kannadaphonserv. This is designed to be a server
which will accept utf text from a client, convert in to
phonetics, and pass it on to the speech engine server, ./synthserv.
An example of a UTF kannada text is in tdemos/kannada/utf.
An example of a client which will sent UTF to  kannadaphonserv
is in simputer/client. See README in simputer/ to find out
how to invoke this client.

4. A convertor which takes roman transliteration of a hindi
text and converts to UTF format, called tdemos/hindi/utfconvert.pl.
The transliteration format is explained in the newutfgen.pl
file itself.

5. A convertor which takes roman transliteration of a kannada
text and converts to UTF format, called tdemos/kannada/utfconvert.pl.
The transliteration format is explained in the newutfgen.pl
file itself.


See doc/README and the READMES in the individual directories
for further details.