TTS Builder 5.0

TTS Builder 5.0 is a powerful tool which allows engineers to create their own Text To Speech Voice in any given Language using High Quality Open Festival Speech Synthesis Research.

A complete step-by-step manual is included. With a simple GUI a voice artist can create his/her own text to speech voice. The retail TTSBuilder documentation converts the heavy technical research oriented into one simple practical task.

Not to mention all required libraries for Windows are included. This is an Research Lab attempt to bring speech synthesis to basic developers.

Major Features:

1. Festival Speech Synthesis Research allows one to create production level Text To Speech Engines
   • Tools
   • • Available : A complete multi-lingual speech synthesis workbench off research
     • Ported to Embedded : Edinburgh Speech Tools Library
     • Pen DSP Library : libsnd.dll
     • CMUdict -- pronunciation dictionary
     • Ported
     • • PenVXI -- VoiceXML browser
       • SALT browser -- finally online!
       • Audio Databases -- AN4, Microphone array, etc
   • Advised TTS for Dictionary Resource:
   • • The DICT Development Group: Clients for the RFC 2229 dictionary protocol.
     • Encyclopedia Britannica: Just like the twenty-book set, but it fits in your web browser.
     • Hypertext Webster Gateway: Search engine spanning multiple dictionaries.
     • IEEE Keywords: List of approved IEEE keywords for indexing publications.
     • Merriam-Webster Online: Perhaps the best on-line dictionary available.
     • Roget's Thesaurus: The all-in-one desktop reference - search the web, a dictionary, and Roget's Thesaurus.
   • Directories possible
   • • 555-1212: look up a telephone area code.
     • AnyWho: look up any address by it's telephone number.
     • CEOExpress: highly informational site catering to business professionals.
     • MapQuest: get directions anywhere in the US.
     • United States Postal Service: look up a zip code.
   • General Databases
   • • CMU ARCTIC, 4 single speaker speech databases with around 1200 phonetically balanced utterances.
     • CMU FAF, 107 paragraphs (15,000 words) of single speaker monologues with interesting prosody. Basic of Aesop's fables and country descriptions in the CIA world fact book.
     • CMU SIN, speech in noise: speech recorded while noise is playing in the speakers ears (and when not).
     • CSTR US KED timit University of Edinburgh's male US TIMIT, 452 phonetically balanced utterances.
   • Limited Domain Databases
   • • Telling the time
     • The current weather (in US)
     • Communicator dialog
   • Diphone Databases
   • • CMU US KAL diphone
     • CSTR UK RAB diphone
   • MBROLA voices and binaries: Check out these MBROLA projects wide range of pre-built diphone databases for many languages and binaries for the mbrola program itself for many platforms.

Enhancements:

• SAPI 5.0,
• GUI 
• Documentation with Installer Creation.