Continuous F0 in the source-excitation generation for HMM-based TTS: Do we need voiced/unvoiced classification?

Latorre, Javier; Gales, Mark J. F.; Buchholz, Sabine; Knill, Kate; Tamura, Masahito; Ohtani, Yamato; Akamine, Masami

doi:10.1109/icassp.2011.5947410

Cited by 19 publications

(12 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Whereas the original SBM method focuses on the log-amplitude spectrum and phase spectrum, we apply SBM to the log-amplitude spectrum and aperiodic component. Here, the aperiodic component is defined as the ratio between harmonic and non-harmonic components in each frequency band of a speech signal [30]. The value 0 means complete periodic and 1 is purely non-periodic.…”

Section: Sub-band Basis Spectrum Modelmentioning

confidence: 99%

See 1 more Smart Citation

Statistical Bandwidth Extension for Speech Synthesis Based on Gaussian Mixture Model with Sub-Band Basis Spectrum Model

Ohtani

Tamura

Morita

2016

IEICE Trans. Inf. & Syst.

Self Cite

View full text Add to dashboard Cite

SUMMARYThis paper describes a novel statistical bandwidth extension (BWE) technique based on a Gaussian mixture model (GMM) and a sub-band basis spectrum model (SBM), in which each dimensional component represents a specific acoustic space in the frequency domain. The proposed method can achieve the BWE from speech data with an arbitrary frequency bandwidth whereas the conventional methods perform the conversion from fixed narrow-band data. In the proposed method, we train a GMM with SBM parameters extracted from full-band spectra in advance. According to the bandwidth of input signal, the trained GMM is reconstructed to the GMM of the joint probability density between low-band SBM and high-band SBM components. Then high-band SBM components are estimated from low-band SBM components of the input signal based on the reconstructed GMM. Finally, BWE is achieved by adding the spectra decoded from estimated high-band SBM components to the ones of the input signal. To construct the full-band signal from the narrow-band one, we apply this method to log-amplitude spectra and aperiodic components. Objective and subjective evaluation results show that the proposed method extends the bandwidth of speech data robustly for the log-amplitude spectra. Experimental results also indicate that the aperiodic component extracted from the upsampled narrow-band signal realizes the same performance as the restored and the full-band aperiodic components in the proposed method. key words: speech enhancement, voice conversion, bandwidth extension, sub-band basis spectrum model, Gaussian mixture model

show abstract

Section: Sub-band Basis Spectrum Modelmentioning

confidence: 99%

“…Thus, the full-band in this paper means 50-11025 Hz. Speech spectra were derived by 1024-point pitch synchronous Fourier transform using our in-house tools [30]. When the SBM parameters were extracted, the warping parameter α was set to 0.35 according to [25].…”

Section: Speaker-independent and Speaker-dependent Bandwidth Extensiomentioning

confidence: 99%

Statistical Bandwidth Extension for Speech Synthesis Based on Gaussian Mixture Model with Sub-Band Basis Spectrum Model

Ohtani

Tamura

Morita

2016

IEICE Trans. Inf. & Syst.

Self Cite

View full text Add to dashboard Cite

show abstract

“…More recently, some work was done using continuous F 0 and it was shown that continuous F 0 improves the perceived naturalness of synthesis (Yu and Young, 2011;Latorre et al, 2011). This was further improved by hierarchical modelling using a continuous wavelet decomposition to separate the different levels of variation in F 0 (Suni et al, 2013).…”

Section: Introductionmentioning

confidence: 99%

Intonation modelling using a muscle model and perceptually weighted matching pursuit

Honnet

Gerazov

Gjoreski

et al. 2018

Speech Communication

View full text Add to dashboard Cite

We propose a physiologically based intonation model using perceptual relevance. Motivated by speech synthesis from a speech-to-speech translation (S2ST) point of view, we aim at a language independent way of modelling intonation. The model presented in this paper can be seen as a generalisation of the command response (CR) model, albeit with the same modelling power. It is an additive model which decomposes intonation contours into a sum of critically damped system impulse responses. To decompose the intonation contour, we use a weighted correlation based atom decomposition algorithm (WCAD) built around a matching pursuit framework. The algorithm allows for an arbitrary precision to be reached using an iterative procedure that adds more elementary atoms to the model. Experiments are presented demonstrating that this generalised CR (GCR) model is able to model intonation as would be expected. Experiments also show that the model produces a similar number of parameters or elements as the CR model. We conclude that the GCR model is appropriate as an engineering solution for modelling prosody, and hope that it is a contribution to a deeper scientific understanding of the neurobiological process of intonation.

show abstract

“…To alleviate the perceived hoarseness, popular solutions include the use of continuous F0 contours either for modeling [14], [20], [21] or for synthesis [19]. In this paper, we directly tackle the problem of F0 estimation in glottalized regions of a speech signal.…”

Section: Introductionmentioning

confidence: 99%

“…Standard HMM-based TTS [2] uses multi-space distribution (MSD) to model and generate discontinuous F0 trajectories [18]. Faulty voicing decisions resulting from the F0 extraction phase will cause the deteriorately trained MSD-HMMs to synthesize voiced frames as unvoiced, resulting in hoarse speech, or to synthesize unvoiced frames as voiced, resulting in buzzy speech [19]. When listening to the output of our baseline HMM-based TTS system, we perceived that although the synthetic speech is highly intelligible, its overall quality is greatly degraded by the hoarseness frequently occurring at syllables bearing a glottalized tone.…”

Section: Introductionmentioning

confidence: 99%

F0 Parameterization of Glottalized Tones in HMM-Based Speech Synthesis for Hanoi Vietnamese

Ninh

Yamashita

2015

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYA conventional HMM-based speech synthesis system for Hanoi Vietnamese often suffers from hoarse quality due to incomplete F0 parameterization of glottalized tones. Since estimating F0 from glottalized waveform is rather problematic for usual F0 extractors, we propose a pitch marking algorithm where pitch marks are propagated from regular regions of a speech signal to glottalized ones, from which complete F0 contours for the glottalized tones are derived. The proposed F0 parameterization scheme was confirmed to significantly reduce the hoarseness whilst slightly improving the tone naturalness of synthetic speech by both objective and listening tests. The pitch marking algorithm works as a refinement step based on the results of an F0 extractor. Therefore, the proposed scheme can be combined with any F0 extractor.

show abstract

Continuous F0 in the source-excitation generation for HMM-based TTS: Do we need voiced/unvoiced classification?

Cited by 19 publications

References 8 publications

Statistical Bandwidth Extension for Speech Synthesis Based on Gaussian Mixture Model with Sub-Band Basis Spectrum Model

Statistical Bandwidth Extension for Speech Synthesis Based on Gaussian Mixture Model with Sub-Band Basis Spectrum Model

Intonation modelling using a muscle model and perceptually weighted matching pursuit

F0 Parameterization of Glottalized Tones in HMM-Based Speech Synthesis for Hanoi Vietnamese

Contact Info

Product

Resources

About