A robust unit selection system for speech synthesis

Abstract

There has been interest for many years in diphone-based speech synthesis and, recently, a rapidly increasing interest in unit selection-based synthesis (as illustrated by interest in the CHATR system). The limits of both systems are well known. While intelligibility is generally very high for diphone-based systems, the resulting signals do not sound completely natural. This happens for several reasons, amongst them the limited number of phone variants present in a typical system, and the cost of concatenating at diphone boundaries. For unit selection synthesis, typically phone-based, it is possible to produce sentences that sound surprisingly natural and intelligible from a large database. However, quality is often not consistent, and the main difficulties appear to be related to selecting acoustically appropriate units from a large database with the correct prosodic characteristics. Typically no prosody modification is done. In an effort to capture the best features of both systems a unit-selection and synthesis algorithm has been devised that allows finer control than the CHATR system (version 0.8), both by applying selective prosody modification and by exercising finer control over the units that get chosen for synthesis. Results of experiments based on this version of unit selection synthesis will be presented.

Phoneme based myanmar text to speech system

“…Many TTS systems proposed by [1][2][3][4][5] have been implemented by using concatenative method based on different speech units and they can generate high quality synthesized speech. For Myanmar language, there has been considerable effort on speech processing in Myanmar natural language processing.…”

Removing linear phase mismatches in concatenative speech synthesis

“…[1]- [4] attempts to avoid or reduce spectral discontinuities in formants and spectral tilt by choosing the acoustic units from a large inventory. The selection is based, among other things, on the minimization of the distance between magnitude spectra (usually represented by cepstrum coefficients, or line spectrum frequencies) from the concatenated acoustic units.…”

Multisyn: Open-domain unit selection for the Festival speech synthesis system

“…The general search method (Hunt and Black, 1996) has been refined (e.g. Conkie, 1999;Taylor, 2000;Bulyko and Ostendorf, 2001) and complemented by other procedures for specific tasks such as limited domain speech synthesis (Black and Lenzo, 2000).…”