130 likes | 273 Views
Numerical Text-to-Speech Synthesis System. Presentation By: Sevakula Rahul Kumar. Presentation Outline. Introduction Background 3 Components of TTS System Text Pre-processing Prosody Concatenation What has been done / Future Work Questions. What is a TTS System?.
E N D
Numerical Text-to-Speech Synthesis System Presentation By: Sevakula Rahul Kumar
Presentation Outline • Introduction • Background • 3 Components of TTS System • Text Pre-processing • Prosody • Concatenation • What has been done / Future Work • Questions
What is a TTS System? Definition: • A system which takes as input a sequence of words and converts them to speech Applications: • Services for the hearing impaired • Reading email aloud Commercial TTS Systems: • Festival • Bell Labs TTS
Fundamental Components TTS System words Text Pre-processing Prosody Concatenation
Numerical Text Pre-Processing • Input • A number (integer) • Output • Identification of each digit with their corresponding values • Objective • Perform the analysis of the number • To identify which all speech signals are to be used from the database and in what order they are to be concatenated.
Numerical Text Pre-Processing (Block Diagram) Digit to Speech Translator with corresponding values NumberConverter Database of speech
Number Converter • The numerals with their textual versions 100 one hundred • Handle fractional and decimal numbers 0.25 point two five • Implementation – Simple Search Algorithm in MATLAB using string function – ‘strcpy’
Concatenation • Concatenating arrays of sampled data of the required speech signals • Converting the resultant sampled data again into speech. Individual Digit and position Complete Number
Acoustic Manipulation - MATLAB • Recognizes wave files (.WAV) • Load or read, write • Vast array of signal processing tools • Built-in functions • Ease of debugging
Progress Till First Minor • Work Completed / Current Status • Partial Text Pre-processing for any integer where individual digit is recognised one after another, without considering their value in the entire number. • Since it is a number, no prosodic manipulation was required. • Concatenation. • Work To Be Done • Complete Text Pre-processing. • Speech synthesis of any sized floating point numbers.
Progress Since First Minor • Work Completed • Complete Text Pre-processing. • Speech Synthesis of Floating Point Numbers • Work To Be Done • Speech synthesis of any sized floating point numbers.