Boris Doval

On the use of the derivative of electroglottographic signals for characterization of nonpathological phonation

Electroglottography is a common method for providing noninvasive measurements of glottal activity. The derivative of the electroglottographic signal, however, has not attracted much attention, although it yields reliable indicators of glottal closing instants. The purpose of this paper is to provide a guide to the usefulness of this signal. The main features that are to be found in this signal are presented on the basis of an extensive analysis of a database of items sung by 18 trained singers. Glottal opening and closing instants are related to peaks in the signal; the latter can be used to measure glottal parameters such as fundamental frequency and open quotient. In some cases, peaks are doubled or imprecise, which points to special (but by no means uncommon) glottal configurations. A correlation-based algorithm for the automatic measurement of fundamental frequency and open quotient using the derivative of electroglottographic signals is proposed. It is compared to three other electroglottographic-based methods with regard to the measurement of open quotient in inverse-filtered derived glottal flow. It is shown that agreement with the glottal-flow measurements is much better than most threshold-based measurements in the case of sustained sounds.

Estimation of fundamental frequency of musical sound signals

In order to estimate the fundamental frequency (fO) of pseudoperiodical sounds with a wide band of possible fO, a theoretical model based on a maximum likelihood for fO is proposed. The model is simplified to make it fast enough for extensive tests. The resulting algorithm is tested on musical speech sounds. As a musical application, an instrument follower based on the algorithm and operating in real time is implemented.<<ETX>>

Zeros of Z-transform representation with application to source-filter separation in speech

We propose a new spectral representation called the zeros of z-transform (ZZT), which is an all-zero representation of the z-transform of the signal. We show that separate patterns exist in ZZT representations of speech signals for the glottal flow and the vocal tract contributions. A decomposition method for source-tract separation is presented based on ZZT. The ZZT-decomposition consists in grouping the zeros into two sets, according to their location in the z-plane. This type of decomposition leads to separating glottal flow contribution (without a return phase) from vocal tract contribution in the z domain.

Spectral correlates of glottal waveform models: an analytic study

This paper deals with the spectral representation of the glottal flow. The LF and the KLGLOTT88 models of the glottal flow are studied. We compute analytically the spectrum of the LF-model. Then, formulas are given for computing spectral tilt and amplitudes of the first harmonics as functions of the LF-model parameters. We consider the spectrum of the KLGLOTT88 model. It is shown that this model can be modeled in the spectral domain by an all-pole third-order linear filter. Moreover, the anticausal impulse response of this filter is a good approximation of the glottal flow model. Parameter estimation seems easier in the spectral domain. Therefore our results can be used for modification of the (hidden) glottal flow characteristic of natural speech signals, by processing directly the spectrum, without needing time-domain parameter estimation.

Just noticeable differences of open quotient and asymmetry coefficient in singing voice

This study aims to explore the perceptual relevance of the variations of glottal flow parameters and to what extent a small variation can be detected. Just Noticeable Differences (JNDs) have been measured for three values of open quotient (0.4, 0.6, and 0.8) and two values of asymmetry coefficient (2/3 and 0.8), and the effect of changes of vowel, pitch, vibrato, and amplitude parameters has been tested. Two main groups of subjects have been analyzed: a group of 20 untrained subjects and a group of 10 trained subjects. The results show that the JND for open quotient is highly dependent on the target value: an increase of the JND is noticed when the open quotient target value is increased. The relative JND is constant: deltaOq/Oq = 14% for the untrained and 10% for the trained. In the same way, the JND for asymmetry coefficient is also slightly dependent on the target value--an increase of the asymmetry coefficient value leads to a decrease of the JND. The results show that there is no effect from the selected vowel or frequency (two values have been tested), but that the addition of a vibrato has a small effect on the JND of open quotient. The choice of an amplitude parameter also has a great effect on the JND of open quotient.

REALTIME AND ACCURATE MUSICAL CONTROL OF EXPRESSION IN SINGING SYNTHESIS

In this paper, we describe a full computer-based musical instrument allowing realtime synthesis of expressive singing voice. The expression results from the continuous action of an interpreter through a gestural control interface. In this context, expressive features of voice are discussed. New real-time implementations of a spectral model of glottal flow (CALM) are described. These interactive modules are then used to identify and quantify voice quality dimensions. Experiments are conducted in order to develop a first framework for voice quality control. The representation of vocal tract and the control of several vocal tract movements are explained and a solution is proposed and integrated. Finally, some typical controllers are connected to the system and expressivity is evaluated.

Phase-based methods for voice source analysis

Voice source analysis is an important but difficult issue for speech processing. In this talk, three aspects of voice source analysis recently developed at LIMSI (Orsay, France) and FPMs (Mons, Belgium) are discussed. In a first part, time domain and spectral domain modelling of glottal flow signals are presented. It is shown that the glottal flow can be modelled as an anticausal filter (maximum phase) before the glottal closing, and as a causal filter (minimum phase) after the glottal closing. In a second part, taking advantage of this phase structure, causal and anticausal components of the speech signal are separated according to the location in the Z-plane of the zeros of the Z-Transform (ZZT) of the windowed signal. This method is useful for voice source parameters analysis and source-tract deconvolution. Results of a comparative evaluation of the ZZT and linear prediction for source/tract separation are reported. In a third part, glottal closing instant detection using the phase of the wavelet transform is discussed. A method based on the lines of maximum phase in the time-scale plane is proposed. This method is compared to EGG for robust glottal closing instant analysis.

Performative Voice Synthesis for Edutainment in Acoustic Phonetics and Singing: A Case Study Using the “Cantor Digitalis”

A real-time and gesture controlled voice synthesis software is applied to edutainment in the field of voice pedagogy. The main goals are teaching how voice works and what makes the differences between voices in an interactive, real-time and audio-visual perspective. The project is based on ”Cantor Digitalis”, a singing vowel digital instrument, featuring an improved formant synthesizer controlled by a stylus and touch graphic tablet. Demonstrated in various pedagogical situations, this application allows for simple and interactive explanation of difficult and/or abstract voice related phenomena, such as source-filter theory, vocal formants, effect of the vocal tract size, voice categories, voice source parameters, intonation and articulation, etc. This is achieved by systematic and interactive listening and playing with the sound of a virtual voice, related to the hand motions and dynamics on the tablet.

Phonetograms of laryngeal source parameters for different vowels and laryngeal mechanisms

This paper explores how the laryngeal source is adjusted by the singer when the sung vowel changes. As the source parameter adjustments may depend on intensity and pitch, comparisons are done through phonetograms computed for each vowel. Distinction is made in the phonetograms between the two main laryngeal mechanisms (M1 and M2). Male and female subjects were recorded producing crescendos and decrescendos from C3 to C5 on /a/, /i/ and /o/ in order to obtain a reduced phonetogram for each vowel and each laryngeal mechanism. Sound, electroglottographic signals and vertical larynx position (VLP) were recorded. Investigations were done on the open quotient and the VLP. The results show a smaller vocal dynamic on /i/ (and in a smaller extent on /o/) in each laryngeal mechanism, and differences on the intensity low limit of the highest part of the M1‐phonetogram. Clear tendencies about the larynx position were measured. They are similar for both laryngeal mechanisms, but differ among singers. The Oq seems to b...