[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: New Java-based software speech synthesizer available

You're confusing two tangled issues when you correctly state
your preference for an open source implementation, and 
the fact that the Viavoice RPMs have not been updated in a

On the other hand, so long as the free-loaders of the world
insist on not paying for anything --and that by the way *was
not* what the Free Software movement is about--
you'll continue to see  the present status quo.
Festival and Festival Lite exist because of years of speech
research by Alan Black and his team at Edinburgh and CMU.

>>>>> "tcross" == tcross  <tcross@northnet.com.au> writes:

    tcross> This is very interesting. As it is an open
    tcross> source software synth, it may be a good
    tcross> replacement for ViaVoice. I like ViaVoice very
    tcross> much, but because we have to rely on a binary
    tcross> only distribution and as it seems new versions
    tcross> are not being released, I fear it may get
    tcross> difficult to support over time. For example,
    tcross> when I wanted to get viavoice running on a
    tcross> debian system recently, I had to install extra
    tcross> C++ libraries which were not part of the
    tcross> standard install because the ones originally
    tcross> installed were a later version and Via Voice
    tcross> needed the earlier version.

    tcross> Of course, the problem with the Java based TTS
    tcross> under linux is identifying the best JVM to
    tcross> use. Its been a while since I have done any
    tcross> java, but the last time I looked, you had JVMs
    tcross> from Java, Blackdown and IBM as well as some
    tcross> open source interpreters etc. This may make
    tcross> things a bit confusing for the novice
    tcross> user. However, it would be great if I could use
    tcross> emacspeak on my windows machine at work.

    tcross> It looks like I may have found an interesting
    tcross> project to tackle when I take my holidays in
    tcross> March!

    tcross> Tim Jason White writes:
    >> The following announcement appeared on the
    >> java-access mailing list. I am forwarding it to the
    >> Emacspeak list because, on the FreeTTS web site at
    >> Sourceforge, the authors mention an Emacspeak demo
    >> (there appears to be an Emacspeak speech server
    >> available). I haven't downloaded any of the software
    >> yet, so these remarks are based purely on the web
    >> page.
    >> Forwarded message
    >> From: Willie Walker <william.walker@SUN.COM> To:
    >> JAVA-ACCESS@JAVA.SUN.COM Subject: Sun Microsystems
    >> Laboratories releases an open source speech
    >> synthesizer Date: Wed, 19 Dec 2001 17:25:17 -0500
    >> Greetings!
    >> It is my pleasure to announce that the Sun
    >> Microsystems Laboratories Speech Group has made its
    >> FreeTTS (http://freetts.sourceforge.net/) speech
    >> synthesis engine available via open source through a
    >> BSD-style license.  The engine is written entirely in
    >> the Java(tm) programming language and provides
    >> partial support for the synthesis portion of the Java
    >> Speech API 1.0 specification.
    >> You can read more about this project in an article on
    >> http://java.sun.com:
    >> http://java.sun.com/features/2001/12/flite.html
    >> An excerpt from the article is as follows:
    >> "Researchers from Sun Microsystems Laboratories in
    >> Burlington, Massachusetts have created an open source
    >> speech synthesis engine written entirely in the
    >> Java(tm) programming language. This high-performance
    >> software converts text to speech. You type it; your
    >> workstation speaks it. And the whole world benefits.
    >> Willie Walker, Paul Lamere, and Philip Kwok combined
    >> the Festival Speech Synthesis System, with its robust
    >> architecture, and the Flite engine, with its succinct
    >> algorithms, to create FreeTTS, a synthesizer that
    >> delivers both power and flexibility.
    >> The team ported Flite, programmed in C, and Festival,
    >> written in C++ and Scheme, to the Java programming
    >> language. FreeTTS generated intelligible speech four
    >> weeks after researchers wrote the first line of
    >> code. But even with such a short development time,
    >> the team did not compromise results. FreeTTS
    >> outperforms both original applications, executing
    >> nearly four times faster than Flite in some
    >> environments."
    >> For the Sun Labs Speech Group,
    >> Willie Walker, Manager and Principal Investigator
    >> -----------------------------------------------------------------------------
    >> To unsubscribe from the emacspeak list or change your
    >> address on the emacspeak list send mail to
    >> "emacspeak-request@cs.vassar.edu" with a subject of
    >> "unsubscribe" or "help"

    tcross> -- Tim Cross mailto: tcross@northnet.com.au
    tcross> phone: +61 2 6772 5973 mobile: 0412 969193
    tcross> ------------------------------------------------------------------------

    tcross> -----------------------------------------------------------------------------
    tcross> To unsubscribe from the emacspeak list or change
    tcross> your address on the emacspeak list send mail to
    tcross> "emacspeak-request@cs.vassar.edu" with a subject
    tcross> of "unsubscribe" or "help"

Best Regards,

Email:  raman@cs.cornell.edu
WWW: http://www.cs.cornell.edu/home/raman/             
PGP:    http://www.cs.cornell.edu/home/raman/raman.asc 

To unsubscribe from the emacspeak list or change your address on the
emacspeak list send mail to "emacspeak-request@cs.vassar.edu" with a
subject of "unsubscribe" or "help"

Emacspeak Files | Subscribe | Unsubscribe | Search