A Log Domain Pulse Model for Parametric Speech Synthesis

Abstract

Most of the degradation in current Statistical Parametric Speech Synthesis (SPSS) results from the form of the vocoder. One of the main causes of degradation is the reconstruction of the noise. In this article, a new signal model is proposed that leads to a simple synthesizer, without the need for ad-hoc tuning of model parameters. The model is not based on the traditional additive linear source-filter model, it adopts a combination of speech components that are additive in the log domain. Also, the same representation for voiced and unvoiced segments is used, rather than relying on binary voicing decisions. This avoids voicing error discontinuities that can occur in many current vocoders. A simple binary mask is used to denote the presence of noise in the time-frequency domain, which is less sensitive to classification errors. Four experiments have been carried out to evaluate this new model. The first experiment examines the noise reconstruction issue. Three listening tests have also been carried out that demonstrate the advantages of this model: comparison with the STRAIGHT vocoder; the direct prediction of the binary noise mask by using a mixed output configuration; and partial improvements of creakiness using a mask correction mechanism.

A Comparison of Recent Waveform Generation and Acoustic Modeling Methods for Neural-Network-Based Speech Synthesis

“…This work used a conventional vocoder called WORLD [10] as the baseline. It then included a phase-recovery technique [12], a waveform synthesizer based on a log-domain pulse model [11], and a Wavenet-based vocoder for comparison. Complex-valued approaches may be included in future work.…”

Frequency Domain Variants of Velvet Noise and Their Application to Speech Processing and Synthesis

“…How to analyze and generate the random component for synthetic voice has been a difficult problem [5,7,11,12]. In addition to this difficulty in analysis and synthesis, auditory perception introduces another difficulty.…”

“…Group delay manipulation used in legacy-STRAIGHT was successful for reducing this impression [4]. The log domain pulse model (LDPM) also uses phase manipulation [7]. However, such manipulation results smearing of the signal in the time domain.…”

GlotNet—A Raw Waveform Model for the Glottal Excitation in Statistical Parametric Speech Synthesis

“…Meanwhile, recent signal processing methods for vocoding have improved the synthetic speech quality. These techniques include sourcefilter models [7], [8], sinusoidal harmonic-plus-noise models [9], advanced aperiodicity models [10], [11], and direct modeling of the magnitude and phase spectra [12]. Furthermore, the ongoing emergence of neural network waveform generation models, i.e.…”