Authors : Mohammad Moniruzzaman , Faisal Ahmed and Mohammad Mahmudur Rahman
Abstract: Text Coding (TC) is a voice compression algorithm of very high compression rate. In this coding, speech is recognized and broken into phonemes. The phonemes are encoded into text and transmitted. In the receiver the encoded text is decoded into speech. Text Coding takes very little bandwidth as it transmits text message instead of compressed voice. Consistent pronunciation is not a goal, while predictable pronunciation is. Using the phoneme set, an application developer may guarantee a minimal pronunciation, but not the exact expression. In Grammatical Coding (GC) system, words of text are identified by grammar rules. Each word is then encoded with a predefined number. The encoded number is then again compressed with suitable compression algorithm and transmitted by the communication media. The compressed data is at first decompressed in the receiver. Then from the encoded numbers, the words are decoded by Reverse Grammatical Coding (RGC), which is opposite to Grammatical Coding. Finally the decoded words are converted to speech by Text To Speech (TTS) engine.
Mohammad Moniruzzaman , Faisal Ahmed and Mohammad Mahmudur Rahman , 2006. Text and Grammatical Coding: Voice Compression Algorithm with High Compression Rate. Asian Journal of Information Technology, 5: 536-539.