Automatic marking of Punjabi syllables boundaries in a sound file

Automatic Speech segmentation is one of the foremost important field for controlling the simulation of speech without intervening of human being. Segmenting the continuous speech signal according to the phonetic transcription is a fundamental task in any voice activated system. Manual segmentation is tedious, time consuming and error prone. Automatic segmentation is not faultless, but it is inherently consistent and results are reproducible. In spite of the endless efforts made in this field, there is no Speech Segmentation System in the market for Punjabi on large vocabulary, though speaker independent systems are available for the languages like English, Tamil. It is a well known fact Syllable based systems perform better than phoneme or word based systems. In our work, we are presenting an effective method for segmentation of Punjabi Speech into syllable like Basic units for ASR systems.

I Introduction

Speech is the fastest and natural mode of communication for people as giving dictation, querying database of information retrieval system or generally giving commands to computer or other device. For decades human beings have been dreaming of an “intelligent machine” which can master the natural speech. In its simplest form, this machine should consist of two subsystems, namely automatic speech recognition (ASR) and speech understanding (SU). The goal of ASR is to transcribe natural speech while SU is to understand the meaning of the transcription. Recognizing and understanding a spoken sentence is obviously a knowledge-intensive process which must take into account all variable information about the speech communication process, from acoustics to semantics and pragmatics.