Formant Tuning Research Articles

Vowel and formant representation in the human auditory speech cortex

Vowels, a fundamental component of human speech across all languages, are cued acoustically by formants, resonance frequencies of the vocal tract shape during speaking. An outstanding question in neurolinguistics is how formants are processed neurally during speech perception. To address this, we collected high-density intracranial recordings from the human speech cortex on the superior temporal gyrus (STG) while participants listened to continuous speech. We found that two-dimensional receptive fields based on the first two formants provided the best characterization of vowel sound representation. Neural activity at single sites was highly selective for zones in this formant space. Furthermore, formant tuning is adjusted dynamically for speaker-specific spectral context. However, the entire population of formant-encoding sites was required to accurately decode single vowels. Overall, our results reveal that complex acoustic tuning in the two-dimensional formant space underlies local vowel representations in STG. As a population code, this gives rise to phonological vowel perception.

Computer simulation of vocal tract resonance tuning strategies with respect to fundamental frequency and voice source spectral slope in singing.

A well-known concept of singing voice pedagogy is "formant tuning," where the lowest two vocal tract resonances ( fR1, fR2) are systematically tuned to harmonics of the laryngeal voice source to maximize the level of radiated sound. A comprehensive evaluation of this resonance tuning concept is still needed. Here, the effect of fR1, fR2 variation was systematically evaluated in silico across the entire fundamental frequency range of classical singing for three voice source characteristics with spectral slopes of -6, -12, and -18 dB/octave. Respective vocal tract transfer functions were generated with a previously introduced low-dimensional computational model, and resultant radiated sound levels were expressed in dB(A). Two distinct strategies for optimized sound output emerged for low vs high voices. At low pitches, spectral slope was the predominant factor for sound level increase, and resonance tuning only had a marginal effect. In contrast, resonance tuning strategies became more prevalent and voice source strength played an increasingly marginal role as fundamental frequency increased to the upper limits of the soprano range. This suggests that different voice classes (e.g., low male vs high female) likely have fundamentally different strategies for optimizing sound output, which has fundamental implications for pedagogical practice.

Effects of Resonant Voice Therapy on Perceptual and Acoustic Source and Tract Parameters – A Preliminary Study on Indian Carnatic Classical Singers

The aim of the study was to examine the effects of resonant voice therapy (RVT) on the vocal resonance of trained Carnatic singers. The specific objectives were to evaluate the effects of resonant voice therapy on the auditory perceptual judgments and acoustic source and tract parameters before and after RVT on phonation and sung voice samples. Six vocally healthy trained Carnatic singers, three males and three females aged 18-25 years (M=23; S.D=2.09) participated in the study. All the participants were assigned to a 21-days-long Resonance Voice Therapy (RVT) training program. The participants' pre and post training phonation and sung samples were subjected to auditory perceptual analysis and acoustic analysis. The results revealed that the post training auditory perceptual ratings of the phonation task showed a statistically significant difference from the pre training scores (Z= 2.35; P=0.019). While for the singing task, the post training perceptual ratings were not significantly different from the pre training perceptual rating scores (Z= 2.66; P=0.08). A significant difference was observed between the pre and post training values for all the measured acoustic parameters of the phonation task. In singing task, though the fundamental frequency, third and fourth formant frequencies showed no significant difference in the pre and post training conditions (P > 0.05), the parameter of- difference between the first formant frequency and the fundamental frequency showed a significant decrease (P=0.028). The effects of resonant voice production led to a high vocal economy, as evidenced from the improved source and filter acoustic parameters. Indication for formant tuning through vocal tract modifications, probably an enlarged pharyngeal area resulting in increased resonant voice quality in both phonation and singing tasks, is inferred from these results.

Svan Funeral Dirges (Zär): Language-Music Relation and Phonetic Properties

This paper reports on the results of a follow-up study to Scherbaum and Mzhavanadze (2020) and Mzhavanadze and Scherbaum (2020), which jointly describe the acoustical and musicological properties, respectively, of a new collection of field recordings of three-voiced Svan funeral dirges, known as zär in Svan and zari in Georgian. The focus of the present work is on language–music relation and phonetic properties. It was motivated by the pioneering work of Bolle Zemp (1994; 1997; 2001), who for the first time examined this topic from both an ethnomusicological and a linguistic perspective. We revisit some of Bolle Zemp’s observations and assumptions with a substantially expanded (by a factor of 11) data set and by using computational phonetic analysis tools not available at the time of her study. Her observation of correlations between pitch, duration, and timbre features related to the interjection of woj (wai) is consistent with our analysis for only some of the singers. Therefore, rather than assigning a general semantic meaning to these correlations, we offer an alternative interpretation as a natural consequence of the formant tuning techniques, which we found employed in the vocal production of some of the singers.

Amplitude Effects of Vocal Tract Resonance Adjustments When Singing Louder

In the literature on vocal pedagogy we may find suggestions to increase the mouth opening when singing louder. It is known that sopranos tend to sing loud high notes with a wider mouth opening which raises the frequency of the first resonance of the vocal tract (fR1) to tune it close to the fundamental. Our experiment with classically trained male singers revealed that they also tended to raise the fR1 with the dynamics at pitches where the formant tuning does not seem relevant. The analysis by synthesis showed that such behaviour may contribute to the strengthening of the singer's formant by several dB-s and to a rise in the centre of spectral gravity. The contribution of the fR1 raising to the overall sound level was less consistent. Changing the extent of the mouth opening with the dynamics may create several simultaneous semantic cues that signal how prominent the produced sound is and how great the physical effort by the singer is. The diminishing of the mouth opening when singing piano may also have an importance as it helps singers to produce a quieter sound by increasing the distance between the fR1 and higher resonances, which lowers the transfer function of the vocal tract at the relevant spectral regions.

Vocal Feminization for Transgender Women: Current Strategies and Patient Perspectives.

Voice feminization for transgender women is a highly complicated comprehensive transition process. Voice feminization has been thought to be equal to pitch elevation. Thus, many surgical procedures have only focused on pitch raising for voice feminization. However, voice feminization should not only consider voice pitch but also consider gender differences in physical, neurophysiological, and acoustical characteristics of voice. That is why voice therapy has been the preferred choice for the feminization of the voice. Considering gender difference of phonatory system, the method for voice feminization consists of changing the following four critical elements: fundamental frequency, resonance frequency related to vocal tract volume and length, formant tuning, and phonatory pattern. Voice feminizing process can be generally divided into non-surgical feminization and surgical feminization. As a non-surgical procedure, feminization voice therapy consists of increasing fundamental frequency, improving oral and pharyngeal resonance, and behavioral therapy. Surgical feminization usually can be achieved by external approach or endoscopic approach. Based on three factors (length, tension and mass) of vocal fold for pitch modulation, surgical procedure can be classified as one-factor, two-factors and three-factors modification of vocal folds. Recent systematic reviews and meta-analysis studies have reported positive outcomes for both the voice therapy and voice feminization surgery. The benefits of voice therapy, as it is highly satisfactory, mostly increase vocal pitch, and are noninvasive. However, the surgical voice feminization of three-factors modification of vocal folds is also highly competent and provides a maximum absolute increase in vocal pitch. Voice feminization is a long transition journey for physical, neurophysiological, and psychosomatic changes that convert a male phonatory system to a female phonatory system. Therefore, strategies for voice feminization should be individualized according to the individual’s physical condition, the desired change in voice pitch, economic conditions, and social roles.

Control of vocal loudness in singing

Vocal loudness in sones is quantified on the basis of changes in spectral slope and harmonic tuning in a range of singing fundamental frequencies of 125 Hz to 1000 Hz. Spectral slope of the mouth output pressure is bracketed in the range of -3 dB/octave to -12 dB/octave to reflect a typical glottal spectrum from breathy to pressed adduction. To approximate formant tuning of harmonics, the SPL level of the first three harmonics is raised by 10, 20, and 30 dB. It is shown that spectral slope change is more effective in increasing vocal loudness than tuning a single harmonic with a vocal tract resonance. Some applications to amplified and unamplified vocal production and respective training are given.

On Exploring Vocal Ornamentation in Byzantine Chant

A special vocal ornament in Byzantine chant (BC), the single cycle ornamentation structure (SCOS), is defined and compared with the vibrato with respect to its time (rate, extent) and spectral (slope [SS], relative speaker's formant [SPF] level, formant frequencies [Fi] and bandwidths [Bi], and noise-to-harmonics ratio [NHR]) characteristics. This is a comparative study between the vocal ornaments of SCOS and vibrato, of which time and spectral acoustic parameters were measured, statistically analyzed, and compared. From the same hymn recordings chanted by four chanters, the SS, SPF level, FFi, FBi, and NHR difference values between the vocal ornament and its neighbor steady note, and the rate and extent, were compared with those of vibrato. The mean extent values for SCOS were found to be almost double the corresponding values for vibrato, and the rate of SCOS tends to be different from the rate of vibrato. The difference values of: 1) the NHR, 2) the spectral slope, and 3) the SPF level, between the vocal ornament and its neighbor steady note were found to be: 1) higher for SCOS, 2) mainly lower for SCOS, and 3) lower for SCOS, respectively. No significant differences were detected for the FFi and FBi. The FF1 differences tend to be negative in both ornaments indicating a formant tuning effect. A new vocal ornament (SCOS) in BC is studied, of which the extent, NHR (HNR), the spectral slope, and the SPF level are different compared to those of vibrato.

The Perception of Formant Tuning in Soprano Voices

At the upper end of the soprano range, singers adjust their vocal tract to bring one or more of its resonances (Rn) toward a source harmonic, increasing the amplitude of the sound; this process is known as resonance tuning. This study investigated the perception of (R1) and (R2) tuning, key strategies observed in classically trained soprano voices, which were expected to be preferred by listeners. Furthermore, different vowels were compared, whereas previous investigations have usually focused on a single vowel. Listeners compared three synthetic vowel sounds, at four fundamental frequencies (f0), to which four tuning strategies were applied: (A) no tuning, (B) R1 tuned to f0, (C) R2 tuned to 2f0, and (D) both R1 and R2 tuned. Participants compared preference and naturalness for these strategies and were asked to identify each vowel. The preference and naturalness results were similar for /ɑ/, with no clear pattern observed for vowel identification. The results for /u/ showed no clear difference for preference, and only slight separation for naturalness, with poor vowel identification. The results for /i/ were striking, with strategies including R2 tuning both preferred and considered more natural than those without. However, strategies without R2 tuning were correctly identified more often. The results indicate that perception of different tuning strategies depends on the vowel and perceptual quality investigated, and the relationship between the formants and (f0). In some cases, formant tuning was beneficial at lower f0s than expected, based on previous resonance tuning studies.

Use of H1-H2 to quantify formant tuning for notes and portions of the vibrato cycle in the second passaggio of a professional female singer

Singing voice pedagogy emphasizes that an acoustic change occurs in standard, classical, techniqued soprano voice between the musical notes D5♮ (587 Hz) and F5♮ (698 Hz). For low vowels, this involves a transition from tuning of the second resonance to the second harmonic (F2/H2) to tuning of the first resonance to the fundamental (F1/F0). In this single-subject study, we quantified the acoustics of this transition as the amplitude difference between the first and second harmonics (H1-H2). Results showed a clear and substantial change from negative to positive H1-H2 values at a pivot point between E5♭ and E5♮, implying the resonance tuning. Non-techniqued singing, with the same singer, showed no such change. F0 fluctuation (vibrato) of ±90 cent at the pivot point resulted in positive H1-H2 values at vibrato maxima and negative ones at vibrato minima. Additionally, H1-H2 values were consistently higher at vibrato maxima than minima throughout the transition area. Potential explanations for the latter result are: (i) vocal tract resonances are located just above the sung F0, or (ii) the vibrato cycle is accompanied by an articulatory change, possibly laryngeal movement. This illustrates the intricacies of formant tuning and suggests future possibilities for numerical assessment of vocal technique.

Formant Tuning and Feedback in the Male Passaggio

It has been suggested that traversing the male (secondo) passaggio requires two important adjustments. When singing up the scale, first of all, the second harmonic (H2) needs to pass over the first formant (F1). After that, the timbre of the voice takes on a different, slightly "darker" quality. This is the pitch where, in singers' jargon, the voice reaches secondo passaggio. Above secondo passaggio, in the optimal arrangement, the second formant (F2) is tuned close to one of the higher harmonics, or, sometimes alternatively, the singer's formant cluster induces a dominant resonance in the approximate range of 2.4-3.4 kHz. These two adjustments together produce the typical sound of the classical male upper voice. In this study, we have investigated the choices individual singers make while negotiating the passaggio and the effect of feedback from the vocal tract to the voice source during this maneuver. Electroglottograph (EGG) and microphone signals were recorded of nine male singers (five tenors and four baritones) using VoceVista. Inverse filtering was performed on the microphone signals, using the Sopran/DeCap application, to reveal the shape of the glottal airflow pulses.

Resonatory Properties in Professional Tenors Singing Above the Passaggio

Introduction: The question of formant tuning in male professional voices has been a matter of discussion for many years. Material and Methods: In this study four very successful Western classically ...

Speech Watermarking Method Based on Formant Tuning

Speech Watermarking Method Based on Formant Tuning

Formant and Voice Source Properties in Two Male Kunqu Opera Roles: A Pilot Study

Objective: This investigation analyzes flow glottogram and electroglottogram (EGG) parameters as well as the relationship between formant frequencies and partials in two male Kunqu Opera roles, Colorful face (CF) and Old man (OM). Participants and Methods: Four male professional Kunqu Opera singers volunteered as participants, 2 singers for each role. Using inverse filtering of the audio signal flow glottogram parameters and formant frequencies were measured in each note of scales. Two EGG parameters, contact quotient (CoQ) and speed quotient, were measured. Results: Formant tuning was observed only in 1 of the OM singers and appeared in a pitch range lower than the passaggio range of Western male opera singers. Both the CF and the OM role singers showed high CoQ values and low values of the normalized amplitude quotient in singing. For 3 of the 4 singers CoQ and the level difference between the first and second partials showed a positive and a negative correlation with fundamental frequency (F0), respectively. Conclusions: Formant tuning may be applied by a singer of the OM role, and both CF and OM role singers may use a rather pressed type of phonation, CF singers more than OM singers in the lower part of the pitch range. Most singers increased glottal adduction with rising F0.

Formant Tuning Strategies in Professional Male Opera Singers

The term "formant tuning" is generally used for the case that one of the lowest formant frequencies coincides with the frequency of a source spectrum partial. Some authors claim that such coincidence is favorable and belongs to the goals of classical opera voice training, whereas other authors have found evidence for advising against it. This investigation analyzes the relationships between formant frequencies and partials in professional singers, who sang scales on the vowels /a/, /u/, /i/, and /ae/ in a pitch range including the passaggio, that is, the fundamental frequency range of approximately 300-400Hz, applying either of the two singing strategies that are typically used (1) in classical and (2) in nonclassical singing, respectively. Formant frequencies of each note in the scales were measured by inverse-filtering the acoustic signal. In the classical style, the first formant tended to be lower than in the nonclassical style. Neither the first nor the second formant tended to change systematically between scale tones, such that on some scale tones either or both formants was just below, just above, or right on a spectrum partial. In many cases, singers produced similar spectrum characteristics of the top tones of the scales with different first and second formant frequencies. Regardless of whether the first formant was slightly lower, slightly higher, or right on a partial, the properties of the voice source did not seem to be affected.

Professional male singers’ formant tuning strategies for the vowel /a/

Certain spectrum characteristics have been identified as important for register equalization around the male passaggio, an effect ascribed to formant tuning although descriptions of formant tuning diverge. Eight professional singers sang scales including their passaggio range on different vowels, applying two formant tuning strategies as found in () classical and () non-classical singing. Formant frequencies were measured using inverse filtering. Results revealed differences between the two strategies. For the classical formant tuning, systematic changes of formant frequencies with pitch were observed. For the highest note sung on /a/, was below the second partial and in the vicinity of the third. Similar spectrum characteristics were achieved by different and values between singers.

Formant strategies in professional male singers.

Singers use formant tuning to avoid discontinuities in voice quality between registers. Around the passaggio, a set of spectrum characteristics has been identified as important to achieve this register equalization. Previous studies suggest different formant tuning strategies in male voices. The present investigation attempts to corroborate these suggestions experimentally. Eight professional tenor or baritone singers sang ascending scales on the vowels [a, æ, i, u] ending on pitches F4, F4, or G4 (350, 370, and 392 Hz, respectively) and including their passaggio notes. These scales were sung with two different formant strategies: (a) as in their normal classical singing technique and (b) as in a non-classical, musical theatre style. Formant frequencies were measured using the custom made DECAP inverse-filtering software (GRANQVIST). Results revealed formant tuning differences between the two styles. For the classical singing technique, systematic changes of formant frequencies with pitch were observed. For the back vowels F1 was adjusted to a frequency below the second partial and F2 to the vicinity of the third, confirming previous observations. Formant frequencies were often quite close to a partial. The major spectrum characteristics could be accurately replicated by synthesis applying the source filter theory of voice production.

Team Management of Voice Disorders in Singers

Team Management of Voice Disorders in Singers

Resonant Voice in Acting Students: Perceptual and Acoustic Correlates of the Trained Y-Buzz by Lessac

This study aimed to investigate perceptually and acoustically Lessac's Y-Buzz and sustained productions of Brazilian Portuguese habitual /i/ vowels pre- and posttraining and to verify the presence of formant tuning and its association with the perception of a more resonant voice. The subjects of this study were 54 acting students, 31 female and 23 male, with no voice problems, distributed in seven groups. Each group received four weekly sessions of training. For the pretraining recording, they were asked to sustain the vowel /i/ in a habitual mode three times at self-selected comfortable frequencies and intensity. After training, they repeated the habitual /i/ and also the trained Y-Buzz. Five voice specialists rated how resonant each sample sounded. The fundamental frequency (F(0)), the first four formant frequencies, the distance between the frequencies of F(1) and F(0) were measured, as well as the harmonic frequency (H(2)) frequency and the difference between F(1) and H(2) in the case of male voices (Praat 4.4.33, Institute of Phonetic Sciences, University of Amsterdam, The Netherlands). The trained Y-Buzz was considered more resonant than the habitual /i/ samples, regardless the gender and demonstrated a lowering of the four formant frequencies. F(1) was especially lower in both groups (288Hz-female and 285Hz-male), statistically significant in the female group. The F(1)-F(0) difference was significantly smaller for the female Y-Buzz (52Hz), as well as F(1)-H(2) in the case of the male Y-Buzz (12Hz), suggesting formant tuning. It was not possible to establish association between the perceptual grades and measures F(1)-F(0) or F(1)-H(2).

Tuning phenomena of melodies and resonance frequencies (formants) in infants pre-speech utterances

In a former study [Wermke et al., Med. Eng. Phys. 24, 501–514 (2002)], an increasing tuning between laryngeal (melody) and pharyngeal (resonance frequencies) activity was demonstrated during pre-speech development. This tuning was observed unexpectedly early during development in mitigated cries and earliest non-cry utterances and prepares probably articulation in speech-like vocalizations of older infants. A new retrospective study supported this assumption by comparing tuning processes in cries and early non-cry utterances (N=2500) in two groups (low versus high word production performance at 18 months) of term-born healthy infants (N=20). Additionally, age-matched comparisons were made in 4 cleft-lip-palate-infants treated with a palatal plate. In order to demonstrate the interaction between melody and resonance control we designed a special graphical representation. The tracking function of the resonance frequencies is displayed synchronously to the melody and its harmonics. Resonance frequencies in pre-speech utterances are not yet identical to formant bands associated with speech sounds. The results support the existence of an early active tuning and its relation to later speech. This behavior seems to prepare formant tuning in later speech. Medical applications are seen for infants with disturbances of the vocal tract transfer function, e.g., infants with cleft-lip-palate. [Work supported by DFG.]