An enhanced ABS/OLA sinusoidal model for waveform synthesis in TTS
نویسندگان
چکیده
This paper describes a method for text-to-speech waveform synthesis based on the Analysis-by-Synthesis/Overlap-Add (ABS/OLA) sinusoidal model. This model has been shown in previous work to be a useful framework for pitch and time-scale modi cation of both speech and music signals. This paper explores some extensions of the original ABS/OLA formulation that attempt to overcome speci c artifacts, including a phase dithering approach for unvoiced speech synthesis and an improved pitch modi cation method that compensates for undesirable energy modulation e ects. The implementation of the model within a text-to-speech synthesis (TTS) system is described, and the results of a listener evaluation of the method are discussed. 1. THE ABS/OLA SINUSOIDAL MODEL The Analysis-by-Synthesis Overlap-Add (ABS/OLA) sinusoidal model [1, 2], represents an input signal s[n] by a sum of equal-length, overlapping short-time signal frames sk[n].
منابع مشابه
Speech quality improvement in TTS system using ABS/OLA sinusoidal model
In this paper, we propose a novel unit concatenation and synthesis method using ABS/OLA sinusoidal model. Phase succession is used in the unit synthesis assuming that the pitch onset time of the rst frame in a given unit is the frame center. In the unit concatenation, the phase succession and interpolation of the sinusoid amplitudes via several frames around the concatenation point is utilized....
متن کاملAdvances in Spectral Parameterization for Statistical (HMM-Based) TTS
HMM-based parametric speech synthesis has recently become an alternative to the concatenative TTS approach, especially when low footprint and general speech domain are required. A majority of speech parameterization models used in state-ofthe art HMM TTS systems employ source-filter waveform synthesis schemes. Sinusoidal representation and waveform generation of speech is an alternative to the ...
متن کاملPractical high-quality speech and voice synthesis using fixed frame rate ABS/OLA sinusoidal modeling
This paper describes algorithms developed to apply the Analysis-by-Synthesis/Overlap-Add (ABS/OLA) sinusoidal modeling system to real-time speech and singing voice synthesis. As originally proposed, the ABS/OLA system is limited to unidirectional timescaling, and relies on variable frame length to accomplish time-scale modification. For speech and voice synthesis applications, unidirectional ti...
متن کاملAnalysis-by-synthesis low-rate multimode harmonic speech coding
This paper presents an analysis-by-synthesis multimode harmonic coder (AbS-MHC) that employs new techniques to improve both the speech model accuracy and the parameter estimation robustness in the low rate harmonic coding framework. To improve the speech model accuracy, an enhanced frequency domain transition model is used in conjunction with the sinusoidal model based harmonic coding of voiced...
متن کاملImplementing Real - Time MIDI Music Synthesis Algorithms , ABS / OLA , and SMS for the TMS 320 C 32 DSP
This application report describes a real-time MIDI music synthesis system using a low cost digital signal processor (DSP) such as the Texas Instruments (TITM) TMS320C32 in a PC environment. The system consists of a MIDI device with a MIDI interface, an IBM compatible personal computer, and a TMS320C32 development board where the core of the music synthesis engine resides. The MIDI device genera...
متن کامل