High Quality Sinusoidal Modeling of Wideband Speech for the Purposes of Speech Synthesis and Modification

Chazan, Dan; Hoory, Ron; Sagi, Ariel; Shechtman, Slava; Sorin, Alexander; Shuang, Zhi Wei; Bakis, Raimo

doi:10.1109/icassp.2006.1660161

Cited by 12 publications

(21 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Replacing the original phase with random phase values reduces the nonlinear HASQI values, so the elimination of the TFS is achieved at the cost of reduced envelope fidelity. It would be expected that improved sinusoidal models (Levine et al, 1998;Chazan et al, 2006) would yield better HASQI values, but even so, there appears to be little benefit to this approach in comparison to the other approaches considered in this paper.…”

Section: Resultsmentioning

confidence: 78%

Spectro-temporal envelope changes caused by temporal fine structure modification

Kates¹

2011

The Journal of the Acoustical Society of America

View full text Add to dashboard Cite

The study of speech from which the temporal fine structure (TFS) has been removed has become an important research area. Common procedures for removing TFS include noise and tone vocoders. In the noise vocoder, bands of noise are modulated by the envelope of the speech within each band, and in the tone vocoder the carrier is a sinusoid at the center of each frequency band. Five different procedures for removing TFS are evaluated in this paper: the noise vocoder, a low-noise noise approach in which the noise envelope is replaced by the speech envelope in each frequency band, phase randomization within each band, the tone vocoder, and sinusoidal modeling with random phase. The effects of TFS modification on the speech envelope are evaluated using an index based on the envelope time-frequency modulation. The results show that for all of the TFS techniques implemented in this study, there is a substantial loss in the accuracy of reproduction of the envelope time-frequency modulation. The tone vocoder gives the best accuracy, followed by the procedure that replaces the noise envelope with the speech envelope in each band.

show abstract

Section: Resultsmentioning

confidence: 78%

Spectro-temporal envelope changes caused by temporal fine structure modification

Kates¹

2011

The Journal of the Acoustical Society of America

View full text Add to dashboard Cite

show abstract

“…Speech sinusoidal modeling [1] has been long in the core of many speech generation models, used in a wide range of applications, such as speech coding [1], speech transformation [2] and speech synthesis [3] [4] [5] [6] [7]. It has been successfully applied both in unit-selection [3] [4] and model-based TTS [5] [6] [7].…”

Section: Introductionmentioning

confidence: 99%

“…Stationary unvoiced signals can also be reliably represented by this model, assuming dense enough sampling of the spectrum [1]. The sinusoidal models, having a precise harmonic frequency structure [2] [8], are called Harmonic Models (HMs). They are more simple than general sinusoidal models, as no frequency information besides the pitch is required for the reconstruction.…”

Section: Introductionmentioning

confidence: 99%

“…They are more simple than general sinusoidal models, as no frequency information besides the pitch is required for the reconstruction. HMs are capable of high quality speech reconstruction [2] [8] and transformation [2] [8] [9]. However, the clear drawback of the HMs is their lack of explicit noise modeling despite the large amount of parameters that implicitly represent the noise (e.g.…”

Section: Introductionmentioning

confidence: 99%

“…[2] [9]. Various speech models exist, that extend the sinusoidal modeling with explicit stochastic modeling above certain maximal voicing frequency [2] [10]. Usually a maximal voicing frequency is determined per speech frame to add an ad hoc stochastic component in the high band [2] [10].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Wideband Harmonic Model: Alignment and Noise Modeling for High Quality Speech Synthesis

Shechtman¹,

Sorin²

2016

9th ISCA Workshop on Speech Synthesis Workshop (SSW 9)

View full text Add to dashboard Cite

Speech sinusoidal modeling has been successfully applied to a broad range of speech analysis, synthesis and modification tasks. However, developing a high fidelity full band sinusoidal model that preserves its high quality on speech transformation still remains an open research problem. Such a system can be extremely useful for high quality speech synthesis. In this paper we present an enhanced harmonic model representation for voiced/mixed wide band speech that is capable of high quality speech reconstruction and transformation in the parametric domain. Two key elements of the proposed model are a proper phase alignment and a decomposition of a speech frame to "deterministic" and dense "stochastic" harmonic model representations that can be separately manipulated. The coupling of stochastic harmonic representation with the deterministic one is performed by means of intra-frame periodic energy envelope, estimated at analysis time and preserved during original/transformed speech reconstruction. In addition, we present a compact representation of the stochastic harmonic component, so that the proposed model has less parameters than the regular full band harmonic model, with better Signal to Reconstruction Error performance. On top of that, the improved phase alignment of the proposed model provides better phase coherency in transformed speech, resulting in better quality of speech transformations. We demonstrate the subjective and objective performance of the new model on speech reconstruction and pitch modification tasks. Performance of the proposed model within unit selection TTS is also presented.

show abstract

IBM voice conversion systems for 2007 TC-STAR evaluation

Shuang

Bakis

Qin

2008

Tinshhua Sci. Technol.

View full text Add to dashboard Cite

High Quality Sinusoidal Modeling of Wideband Speech for the Purposes of Speech Synthesis and Modification

Cited by 12 publications

References 4 publications

Spectro-temporal envelope changes caused by temporal fine structure modification

Spectro-temporal envelope changes caused by temporal fine structure modification

Wideband Harmonic Model: Alignment and Noise Modeling for High Quality Speech Synthesis

IBM voice conversion systems for 2007 TC-STAR evaluation

Contact Info

Product

Resources

About