Akademik Veri Yönetim Sistemi
Proceedings of the IEEE 12th Signal Processing and Communications Applications Conference, SIU 2004, Kusadasi, Türkiye, 28 - 30 Nisan 2004, ss.653-656, (Tam Metin Bildiri)
There are many implementations in text-to-speech synthesis for various languages. Geveze software is one of these. The program is based on vocal tract modeling and compresses speech by LPC method. During the synthesis, for each letter of a given word, the nearest combination of the letter sequences within the words used in training is searched and its parameters are used. As in other systems based on vocal tract modeling, a pulse train for voiced sounds generates excitation while a noise signal is used for the unvoiced. The obtained signal is then amplified with a coefficient special to the sound at that instant and finally sent to an IIR filter of which the filter characteristics are determined by LPC coefficients and the digitized waveform of the speech is obtained. During the training, 10 LPC coefficients, 1 gain, and 1 period information for each of 25 ms. windows which follow each other by 10 ms. phase are obtained. During the synthesis, these values change in every 10 milliseconds with the following window's values. After the digital signal at the output of the IIR filter is converted to analog, it also has to be passed through a Low Pass Filter (LPF) in order to smooth the transitioas between windows. After this filtering, the analog signal is ready to be amplified. The objective of this project is to design this system already running on computer as an integrated circuit and if possible, to have a single chip solution so that the chip may have an optimum cost and performance. © 2004 IEEE.