2026-03-11T16:02:39Z https://meral.edu.mm/oai

oai:meral.edu.mm:recid/3822 2021-12-13T03:58:09Z 1582963302567:1597824273898 user-ucsy

Voice Conversion Method for Myanmar Speech Synthesis Tun, Tar Yar Myo Speaker independent speech recognition isimportant for successful development of speechrecognizer in most real world applications. This paperemphasizes speaker independent based on voiceconversion usinga pitch shifting method,PitchSynchronousOver Lap Add Algorithm (PSOLA).Itdepends on detecting the pitch of the signal(fundamental frequency) and changing it according tothe target pitch period using time stretching with PitchSynchronousOver Lap Add Algorithm (PSOLA), thenresampling the signal in order to have the same playrate.The traditional voice activity detection algorithmswhich track only energy cannot successfully identifypotential speech from input because the unwanted partof the speech also has some energy and appears to bespeech. In the proposed system, VAD (Voice ActivityDetection) that calculates energy of high frequencypart separately as zero crossing rate to differentiatenoise from speech is used. 2015-02-05 http://hdl.handle.net/20.500.12678/0000003822 https://meral.edu.mm/records/3822