TTS

From MobileRead
Jump to: navigation, search

TTS is an acronym for Text To Speech which is a voice synthesizing technology used to read electronic text and produced voice output. Also known as "Read Aloud."

Contents

[edit] Overview

TTS creates sound by interpreting electronic text and then synthesizing the voice. There are several different techniques used to create the voice. In most cases the voice has a distinctly computer like sound but some of the newer technologies support very human like voices. TTS can be accomplished by software or hardware.

Synthesized speech can be created by concatenating pieces of recorded speech that are stored in a database. Systems differ in the size of the stored speech units; a system that stores phones or diphones provides the largest output range, but may lack clarity. Systems are not mutually exclusive. For example a diphone system may have a list of keywords that are often mispronounced. When a word in this list is encountered a recorded word is substituted.

Alternatively, a synthesizer can incorporate a model of the vocal tract and other human voice characteristics to create a completely "synthetic" voice output. This provides a human voice that sounds similar to the person used for the vocal tracts.

The Microsoft Speech API (SAPI 5.1) speech engine has a robotic voice by default but can be augmented with the addition of human voices. This is the primary engine used on Windows systems. Speech processors for Windows include: Microsoft Narrator, JAWS and NVDA.

[edit] Voices

To avoid the computer sounding voice you will need to obtain a specific voice implementation that sounds like a human voice. These are available from

These are available in many languages.

There is also "Eloquence" - very efficient synthesized speech also from Nuance.

[edit] eBook Reader support

Current eBook Readers that include TTS support are:

[edit] Unix Speech engines

For variants of Unix including Linux

  • Flite - A small, fast speech synthesis engine
  • Festival - A good, but slow speech synthesizer. Festival Speech Synthesis System Copyright © University of Edinburgh, 1996,1997. All rights reserved.
  • MBROLA - Mbrola related speech synthesizers (English, French, Spanish, German)

[edit] For more information

Personal tools
Namespaces

Variants
Actions
Navigation
MobileRead Networks
Toolbox
Advertisement