31-01-2013, 11:25 AM
Text to Speech Synthesizer
Text to Speech.ppt (Size: 146 KB / Downloads: 38)
Introduction to the Company
This project is done in Capgemini.
Under the guidance of Mr. Srini Kancha, Senior Project Manager.
Capgemini serves industries like Automotive, Consumer Products, Financial Services, Health, Retail etc.
My project comes under consumer products.
Modules
Structure analysis
Process the input text to determine where paragraphs, sentences and other structures start and end. For most languages, punctuation and formatting data are used in this stage.
Text pre processing
Analyze the input text for special constructs of the language. In English, special treatment is required for abbreviations, acronyms, dates, times, numbers, currency amounts, email addresses and many other forms.
Text-to-phoneme conversion
Convert each word to phonemes. A phoneme is a basic unit of sound in a language. US English has around 45 phonemes including the consonant and vowel sounds. Different languages have different sets of sounds (different phonemes).
Java Speech API
The Java Speech API enables developers of speech-enabled applications to incorporate more sophisticated and natural user interfaces into Java applications and applets.
Two core speech technologies are supported through the Java Speech API:
speech recognition
speech synthesis.
Design Goals for the Java Speech API
Provide support for speech synthesizers and for both command-and-control and dictation speech recognizers.
Provide a robust cross-platform, cross-vendor interface to speech synthesis and speech recognition.
Enable access to state-of-the-art speech technology.
Support integration with other capabilities of the Java platform, including the suite of Java Media APIs.
Be simple, compact and easy to learn.
Future Enhancements
With various types of given text the TTS conversion tool will be tested for naturalness and accuracy and examined by linguistic experts to achieve more correct pronunciation. The outcomes of these examinations shall be incorporated to the TTS.