Improving Glottal Waveform Rank-based Glottal Qua
نویسنده
چکیده
Information on the glottal waveform is an important part of many speech applications. However, glottal waveform estimation remains one of the more inexact sciences of speech processing. The work presented here describes an enhancement to a recently presented algorithm by a new technique involving Rank-Based Glottal Quality Assessment (RB-GQA). The basic premise is to investigate potential measures of glottal quality and use these measures to mark the general trends for determining which glottal waveform estimations are better than others. The work presented here is the beginning of a new research initiative to identify robust methods of glottal waveform estimation across genders for use in speaker analysis applications of normal voices (i.e., no voice pathology).
منابع مشابه
Generative Adversarial Network-Based Glottal Waveform Model for Statistical Parametric Speech Synthesis
Recent studies have shown that text-to-speech synthesis quality can be improved by using glottal vocoding. This refers to vocoders that parameterize speech into two parts, the glottal excitation and vocal tract, that occur in the human speech production apparatus. Current glottal vocoders generate the glottal excitation waveform by using deep neural networks (DNNs). However, the squared error-b...
متن کاملGlottal closure and opening detection for flexible parametric voice coding
The knowledge of glottal closure and opening instants (GCI/GOI) is useful for many speech analysis applications. A Pitchsynchronous waveform encoding of voice is one such application. In this paper, a dynamic programming is employed to solve for the global close/open phase segmentation based on the polynomial parametric waveform of the derivative glottal waveform and its quasi-periodicity. Not ...
متن کاملDepression Detection & Emotion Classification via Data-Driven Glottal Waveforms
This doctoral consortium paper outlines the author’s proposed investigation into the use of the voice-source waveform for affective computing. A data-driven glottal waveform representation, previously examined in the authors earlier doctoral studies for its speaker discriminative abilities, is proposed to be studied for both depression detection and emotion recognition, including severity class...
متن کاملGlottal Waveforms for Speaker Inference & A Regression Score Post-Processing Method Applicable to General Classification Problems
Contributions are made along two main lines. Firstly a method is proposed for using a regression model to learn relationships within the scores of a machine learning classifier, which can then be applied to future classifier output for the purpose of improving recognition accuracy. The method is termed r-norm and strong empirical results are obtained from its application to several text-indepen...
متن کاملEffects of Training Data Variety in Generating Glottal Pulses from Acoustic Features with DNNs
Glottal volume velocity waveform, the acoustical excitation of voiced speech, cannot be acquired through direct measurements in normal production of continuous speech. Glottal inverse filtering (GIF), however, can be used to estimate the glottal flow from recorded speech signals. Unfortunately, the usefulness of GIF algorithms is limited since they are sensitive to noise and call for high-quali...
متن کامل