CURRENT ISSUE
Contests
Feature Article
|
|
Issue #207 October 2007
Embedded Speech
Speech Synthesis for Small Applications
by Nicusor Birsan & Ionut Tarsa
Start | Embedding Speech | Speech Synthesis Techniques | Open-Source Project | System Building Blocks | Sound From Luminary Micro | First Big Porting Problem | Synthesizer | Translator | The LM3S811 Speaks | More Speech Applications | Sources & PDF
OPEN-SOURCE PROJECT
Building a text-to-speech synthesis application from scratch is a hard task, especially because of the complexity of the data files. That’s why we found a project that could be ported into an embedded application.
When dealing with a new application domain, it’s always a good idea to find an open-source project as a starting point. After looking around, we eventually found the eSpeak synthesizer, an open-source project with many attractive features. It produces high-quality English speech, it uses a different synthesis method than other open-source TTS engines, it sounds different (it’s clear and easy to listen to), and it can translate text to phoneme codes so it can be adapted as a front end for another speech-synthesis engine. In addition, the synthesizer’s source code is highly portable, so you can use it with different operating systems, and there are various development tools that you can use to produce and tune phoneme data.
Although the project was originally developed for Acorn/RISC OS computers, Jonathan Duddington turned it into an open-source project last year. The new version is an update and a rewrite that includes relaxed memory and processing power constraints. It also includes support for additional languages.
|
