Browse Prior Art Database

Flexible Text-to-Speech Architecture

IP.com Disclosure Number: IPCOM000117381D
Original Publication Date: 1996-Feb-01
Included in the Prior Art Database: 2005-Mar-31
Document File: 2 page(s) / 43K

Publishing Venue

IBM

Related People

deCarmo, L: AUTHOR

Abstract

Disclosed is a Text-to-Speech (TTS) architecture which can incorporate multiple TTS engines to perform off-line TTS synthesis, and which can dynamically determine the output device to receive data. This architecture can be used, for example, to provide TTS conversion to digital audio for playback through an audio adapter, to provide TTS conversion to a file for off-line audio processing, and to provide a means for routing, splitting, and merging TTS audio stream for unique effects.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 89% of the total text.

Flexible Text-to-Speech Architecture

      Disclosed is a Text-to-Speech (TTS) architecture which can
incorporate multiple TTS engines to perform off-line TTS synthesis,
and which can dynamically determine the output device to receive
data.  This architecture can be used, for example, to provide TTS
conversion to digital audio for playback through an audio adapter, to
provide TTS conversion to a file for off-line audio processing, and
to provide a means for routing, splitting, and merging TTS audio
stream for unique effects.

      The Figure is an overview of the TTS architecture 1, together
with devices to which TTS data is routed.  The TTS architecture 1
includes a TTS Manager 2 for determining which TTS conversion engine
3 to load and use for the conversion of text into digital audio in a
PCM (Pulse Code Modulated) format.  The TTS architecture 1 also
includes one or more TTS Conversion Engines 3, one or more TTS Format
Handlers 4 for converting multiple text formats, and a TTS Stream
Processor 5.

      When a logical TTS device is opened, it determines which TTS
Conversion Engine to load through the TTS Manager 2.  The TTS Manager
2 then determines the type of text on input, such as ASCII or Rich
Text, and loads the appropriate Text Format Handler 4.  The TTS
manager 2 then connects to the target output device(s), such as an
Audio Device 6 and/or Digital Audio File 7.  The Text Format Handler
4 converts the input text into a desired format, TTS Conve...