Printer Friendly

Measuring Mozart: a pilot study testing the accuracy of objective methods for matching a song to a singer.


SELECTING APPROPRIATE REPERTOIRE is a primary responsibility for a singing teacher. Astute repertoire selection can help address the technical and musical development of a singer, as well as advance a singer's career in the case of those who are working professionals. Traditionally, repertoire is assigned on an individual basis by carefully considering the singer's age, gender, technical challenges, personality, musicianship, and developmental level, then cross referencing that singer assessment with an evaluation of the vocal, musical, linguistic, and expressive challenges of potential repertoire. (1) A number of print (2) and web (3) resources exist to assist teachers in choosing potential new literature.

In the past, the selection process has depended upon the acquired ability of the singing teacher both to evaluate his or her students and to make judgments about vocal repertoire based upon personal knowledge and close examination of the literature. In recent years, however, the voice range profile (VRP) and voice dosimetry have begun to show promise as objective tools that could assist teachers with the selection process.

Several articles have suggested ways in which the VRP could be used or improved upon for determining the voice classification of performers or for guiding repertoire choices. Emerich, Titze, Svec, Popolo, and Logan compared laboratory VRPs for eight professional actors with speech range profiles of the actors during a dramatic scene recorded in a laboratory and on stage in performance. (4) At times, the actors exceeded their VRP thresholds when in either of the performance conditions. The study highlighted both the utility of the VRP when comparing laboratory values with actual performance voice usage and the shortcomings of the VRP in predicting what repertoire might exceed a performer's capabilities given the emotional content of live performance. Lamarche, Ternstrom, and Hertegard also looked at the VRP as compared to the performance of repertoire, combining a commercial VRP program with a response button that subjects pressed during moments of difficult production. (5) The subjects, all female professionally trained singers, performed three tasks, including the singing of their best aria, using performance-acceptable quality. Subjects also rated how well the button pressings reflected their experiences and usual vocal challenges; this rating included a viewing of a visual graph of the button responses. The investigators' coupling of objective data from the VRP tasks with (a) the singer's self-assessment captured mid-task and (b) the postperformance evaluation of the data and the responses shows significant promise as a means for enhancing singing teachers' guidance of their students. Lamarche, Ternstrom, and Pabon continued testing improvements of the VRP for singers by examining whether VRPs for singers should be limited to physiological measures (without considering performance quality elements), or whether performance abilities should also be measured in some fashion. (6) After testing thirty female subjects, all professionally active singers, they concluded that examining the voice as it is used in performance has clinical importance, and that both types of VRPs should be used for assessing singers. Finally, Herbst, Duus, Jers, and Svec compared the maximum phonation frequency range (MPFR) of amateur choir singers (as gathered during a VRP) with the required pitch range (RPR) for their chosen voice part within a choir, based upon part ranges specified in a commonly used music reference. (7) Differences between the upper and lower boundaries of the singer's MPFR and the boundaries of the chosen voice part's RPR were measured, along with a derived value of the offset of the RPR from the MPFR. The authors concluded that the upper and lower differences could serve as an indicator of the amount of voice use in extended ranges, while the offset measurement could be useful for assessing the alignment of the voice part's music within the singer's overall capabilities. Both measures showed promise for helping guide singers to an appropriate choir part.

Voice dosimetry may also have applications for the assessment of a singer's capabilities and certainly helps quantify possible difficulties presented by various musical pieces and by extracurricular activities outside of performance. Carroll, Nix, Hunter, Emerich, Titze, and Abaza recorded multiple days of three different vocal dose measurements (time, distance, and cycle dose) on seven professional and semiprofessional singers during rehearsals. (8) Results indicated that increased vocal loading episodes coincided with more harsh subjective ratings of voice quality on the day of the load and between 24-72 hours afterward, and that vocal rest periods prior to a loading episode resulted in improved subjective ratings after the load. This project's coupling of objective data on vocal behaviors with subjective ratings of singer fatigue is in some ways similar to Lamarche, Ternstrom, and Hertegard's experiment enhancing the VRP. Schloneger examined the voice use of two collegiate singers before, during, and after a week of opera rehearsals of leading roles in Stravinsky's The Rake's Progress, gathering time and distance doses with a commercial dosimeter. (9) By comparing the objective data with surveys, activity logs, SVHI ratings and laryngeal examinations, he was able to determine under which circumstances (actually outside of opera rehearsals) the singers experienced the highest vocal doses. It would have been highly interesting, however, had the investigator made estimates of the time and cycle doses of the two roles sung by the singers, to compare predicted values with actual performance data. Gaskill and Cowgill, (10) Gaskill, Cowgill, and Many, (11) and Gaskill, Cowgill, and Tinter (12) examined the voice use of various groups of university students through dosimetry, activity logs, self-ratings of voice quality, and clinical examinations. As was also seen in Schloneger's study, in many cases the highest vocal doses occurred outside of singing times. These studies demonstrate the promise of combining dosimetry with qualitative information on voice use for teachers in guiding the behaviors of their students.

A third area of inquiry has sought to measure the demands of vocal literature relative to a singer's capabilities either by examining the musical score and estimating vocal loading or by combining score examination with singer VRPs and dosimetry. Thurmer proposed the "tessiturogram" (alternate spelling--tessituragram) as a means of calculating and displaying in graphic form the percentage of occurrences of pitches in an opera role. (13) This histogram of sung pitch percentages included a note count total for the role, an estimate of the singing time required by the role, and an estimate of the on-stage time for the role. The primary shortcoming of this research was its failure to include a durational element for each sung pitch. Thurmer, however, foresaw the next step in this inquiry, which was to combine the tessituragram with a singer's VRP in order to objectively correlate the vocal demands of specific literature with a singer's vocal capacity. Titze improved upon Thurmer's work by adding to the process the duration of each sung pitch (based upon the written rhythmic values and a metronome setting) and the frequency in Hertz of each pitch. (14) By doing so, time and cycle doses for each note and for each phrase, aria, song, or even stage role could be precisely calculated. Titze then compared these doses against a singer's VRP to determine whether or not the dose-calculated piece might be suitable for the singer. Combining Titze's method of preparing a tessituragram with a subjective evaluation, Hanrahan calculated doses for a number of Mozart tenor arias, then assigned two arias each to five different tenors. (15) One of the arias assigned to each tenor had a tessituragram matching well with that tenor's VRP, while the other aria assigned did not match the VRP. Recordings were made of each singer performing his respective arias; these recordings were then sent to collegiate teachers of singing for evaluation. The external evaluators and the singers themselves responded to a questionnaire about the suitability of the arias for the singers. One final pair of experiments involving tessituragrams, VRPs, and dosimetry was undertaken by Paolillo and Fussi. (16) In the first experiment, "vocal score profiles" or tessituragrams following the Thurmer model were used in combination with VRPs of singers. The investigators compared the pitch-by-pitch dynamic range (collected from the VRP) for singers with tessituragrams of operatic roles, in order to evaluate the relative vocal economy of the role for the tested singers. The second experiment used a commercial dosimeter to gather dose measures of live performances of the leading tenor roles in Verdi's operas Aida and I due Foscari. Some of the dosimetry measures were then compared with Thurmer-type tessituragrams of musical passages and the entire roles.


These previous investigations led the author to develop a pilot study to combine the VRP, the tessituragram (using Titze's method), dose measurements from the performance of a song, the required pitch range of the song (RPR), and the performance range profile (PRP) of the song to answer the following research questions: (1) How closely does a singer's VRP compare with tessituragram estimates of the vocal load of the song, the song's RPR and the song's PRP? (2) How accurate are tessituragram estimates of the vocal load and recovery in a song?

Previous studies suggest that if a singer's VRP and a song or aria's tessituragram are a good match, the vocal work in question has potential to suit that singer well. However, other studies have shown that performers often exceed laboratory VRP values in performance situations, raising questions about how much faith can be put in predictions based on the VRP. Despite this, it was felt that the boundaries of the singer's VRP would not be exceeded in a VRP/PRP comparison of a classical period art song. Furthermore, given the imposed constraint of an art song with little variation in tempo, it was the author's expectation that tessituragram estimates of voicing time and cycle doses and actual values from a performance would be in close agreement with each other. As the notated rests in a musical score do not constitute all the opportunities for a singer's vocal folds to cease vibrating (for example, consider the presence of unvoiced consonants in the text), close agreement between recovery estimates and measurements were not expected.


The song "An Chloe" by Wolfgang Amadeus Mozart, K. 524, in E[flat] major was chosen for the experiment. As this piece is relatively short in length--approximately two minutes--and was written prior to the widespread use of tempo variations such as rubato within a piece, it seemed a perfect choice for testing the accuracy of estimated cycle and time doses. A soubrette soprano (age 47) with a doctoral degree in vocal performance and a number of operatic and oratorio performance credits agreed to serve as the subject for the experiment. She was familiar with the Mozart song prior to participating in this experiment.

Voice Range Profile

A VRP on the vowel /a/ was collected manually from the subject, using a Quest Type II SE-Pro sound level meter (SLM). Prior to the VRP, the SLM was calibrated at 94 dB and 1000 Hz with an external calibrator. Tokens were gathered at 20 cm and 45 degrees below the subject's mouth, and were collected at whole step intervals below and above the range of the song, and at each semitone throughout the range of the song. Pitches were cued from a keyboard tuned to [A.sub.4]= 440 Hz. Measurements were taken in a Whisperroom noise reduction booth (background average SPL level of 48 dB) measuring 3 feet wide by 5 feet long by 7 feet high. The author played the cue pitch for the singer, the singer checked mouth to microphone distance, then sang. The minimum and maximum values for the VRP were read from the SLM's display screen and recorded in a Microsoft Excel file.

Estimation of Voicing Time Dose, Cycle Dose, and Short-Term Recovery Time

The author used Titze's method for estimating the voicing time dose, cycle dose, and short term recovery time for the song. At first, estimating the time sung at each dynamic level was to be included in the protocol, using the same procedure. However, after consulting three different editions, including one edited by American teacher Van Christy (Miami, FL: Belwin-Mills, 1968; publisher no. EL 2652), another published by the German publisher Peters (New York: Peters, n.d.; publisher no. 299a; plate number 9536), and the definitive scholarly source, the Neue Mozart Ausgabe (Kassel: Barenreiter, 1955-1992; Series 3, Werke 8), only the Belwin-Mills version edited by Christy had dynamic markings, and these were in parentheses to indicate they were editorial. Not wanting to assume dynamics where Mozart did not indicate them, any plans for a dynamic level time estimate were dropped.

According to the Neue Mozart Ausgabe, no metronome marking was indicated by the composer; indeed, the metronome was invented after Mozart's lifetime. The author, an experienced singing teacher, and the subject, also an experienced voice teacher and performer, chose a commonly used performance tempo for the song (half note = 60) for the purposes of estimation. Using this tempo, the pitch and time duration of each rhythmic value indicated in the score was recorded and converted into seconds. Notes with fermati were calculated for time doses and cycle doses as having double the written value, which is also a standard performance practice. By doing so, the total time on each musical note or each rest in the entire song could be calculated. The estimated voicing time dose was determined by summing all the sung values. Cycle doses for each pitch were calculated by multiplying the sung frequency in Hertz for each musical note by its duration in seconds. Total estimated cycle dose for the song was calculated by summing all the individual pitch doses. Recovery time was calculated in a similar fashion to the voicing time estimate using the indicated musical rests for duration.

Table 1 details the calculation procedure for the estimated voicing time dose of 98.5 seconds and the estimated cycle dose of 46802 cycles. Table 2 provides the estimated short term recovery occurrences found in the song.

Recording Session of the Song

The same sound booth, SLM, and keyboard used for the VRP were used for the recording session. A music stand covered in cloth was placed in the booth and positioned below and to the right of the SLM microphone. The music to the song was placed on the stand. The singer wore an Iasus model NT3 contact microphone on her neck at the level of her larynx. The throat microphone signal was recorded at a sampling rate of 44.1 K in mono using Goldwave 4.26 software onto the hard drive of an HP Pavillion dv7 laptop computer. The SLM was set 45 degrees and 20 cm to the left of the singer's mouth, although level with it in the vertical plane, due to the presence of the music stand in front and below the singer's mouth. For the session, fast response time and C weighting were used on the SLM. The DC line output from the SLM was sent through a Kay CSL 4500 preamp and A/D converter, then was recorded onto the hard drive of a Dell Optiplex computer at a sampling rate of 44.1 K using Goldwave 5.13 software. The SLM was calibrated at 94 dB and 1000 Hz with the external calibrator. Then both recording sessions (the throat microphone/laptop and the SLM microphone/Dell PC unit) were started on the two computers. A second calibration tone of the same amplitude and frequency was used as an SPL reference at the start of the SLM microphone recording file. The singer was then given a cue pitch of [C.sub.5] (523 Hz) from the keyboard, after which she sang three /a/vowels on [C.sub.5] at low, high, and moderate dynamic levels, standing 20 cm from the SLM, as a test phonation for recording levels. Following this, the first pitch of the song was cued on the keyboard. The soprano, continuing to stand 20 cm from the SLM and wearing the throat microphone, sang the song without accompaniment in the booth, keeping as accurate a tempo as possible throughout all sung phrases and rests. The author stood at a piano outside the booth and followed along on the music score; with the booth door remaining closed, he gently played cue pitches during musical rests to keep the subject on pitch throughout the song.


Dose Measurements

Both the recording file from the throat microphone and the recording file from the SLM's DC output were analyzed using the software program TF32 (2004 version) to extract time and [F.sub.0] traces of the recordings. The traces were exported as text files and transferred into Microsoft Excel 2010. Using an array function in Excel, voicing occurrences, unvoiced segments, phonation time dose, cycle dose, and phonation percentage were calculated for each recording.

Performance Range Profile

The SLM DC output was also separately analyzed using the Kay CSL 4500 software package's frequency and power analysis features. Time/[F.sub.0] data for the whole song was extracted every 20 milliseconds and exported as a text file and transferred into Excel. A 28.1 dB offset between the output dB reading and the actual value was calculated using the second SLM calibration tone on the recorded sample. Using this dB offset, time/SPL values were calculated for the entire song and exported as a text file and transferred into Excel, where the time/SPL and time/[F.sub.0] data were matched. The matching [F.sub.0]/SPL data from the sung performance were then plotted and compared to the singer's VRP.


VRP and Tessituragram Comparison

Figure 1 displays the singer's dynamic range with respect to musical pitch (taken from her VRP) in comparison with the voicing time and cycle dose estimates for each pitch of the song in graphic fashion. The soprano demonstrated a substantial dynamic range for the pitches of the song; the difference between her minimum and maximum SPL on any single note used in the song ranged from 24-40 dB. Assuming a 3 dB difference between each typically used dynamic level (pianissimo, piano, mezzo piano, mezzo forte, forte, fortissimo), or a 15 dB change from the softest to loudest tone, (17) the singer would still have a substantial dynamic range "reserve" to draw upon for expressive usage in performance, indicating that the song should be at least physiologically a comfortable choice for this singer. Highest time dose values on [E[flat].sub.4], [G.sub.4], [B[flat].sub.4], and [E[flat].sub.5] coincided with dynamic ranges of 40, 24, 27, and 29 dB, respectively. That these pitches had the highest time doses is not surprising, given the tonality of the piece is E[flat]major, and these pitches outline the tonic chord; furthermore, [B[flat].sub.4], the pitch with the highest time and cycle doses, serves as the harmonically important 5 th scale degree in E[flat] major and as the root of the dominant triad (B[flat] major). Finally, the figure demonstrates the impact of increasing [F.sub.0] on cycle dose measurements. The ratio of the time doses for [B[flat].sub.4] (25 s) and [E[flat].sub.5] (13 s) is 1.92 to 1, while the ratio of the same notes' predicted cycle doses is 1.44 to 1 (11654 to 8089 cycles). The higher [F.sub.0] increases the "cost" of the note in terms of vibration exposure to the singer versus the time the tissue is in vibration.

VRP and RPR Comparison

Using the method of Herbst, Duus, Jers, and Svec described in the introduction, the soprano's VRP was compared with the song's RPR. The soprano's range was from [E[flat]sub.3]-[D.sub.6], or 35 ST. The song's RPR was from [D.sub.4]-[A[flat]sub.5], or 19 ST. According to their method, the subject's lower reserve (LR, the number of semitones between the singer's lowest note and the lowest note required by the score) = 11, her upper reserve (UR, the number of semitones between the singer's highest note and the highest note required by the piece) = 6, and her derived tessitura shift, or TS = 2.5. All of these values match typical findings for sopranos by Herbst's team, and indicate that the song sits comfortably within the upper area of the singer's range without challenging the extremes of her range.

VRP and PRP Comparison

Figure 2 is an overlay of the singer's VRP (square and triangle shaped data points) with a plot of the F0/SPL data from the song performance (diamonds). Contrary to expectations, and in agreement with the findings of the NCVS team of Emerich, Titze, Svec, Popolo, and Logan, the performance of the song exceeded the values recorded in the VRP, especially from 370 Hz to 740 Hz. There are a number of possible explanations for why this occurred. As noted in the methods section above, the VRP was performed manually by the author, cueing pitches on a keyboard and reading values off the SLM screen while the singer sang, while the song performance values for [F.sub.0] and SPL were gathered from the DC output of the SLM, a much more sensitive and objective method. A second explanation, in agreement with what the NCVS researchers found, is that the musical and emotional context of performing elicits a greater dynamic range than does a laboratory task. Furthermore, the performance was on a variety of consonants and vowels, while the VRP was on the /a/ vowel only. As song performance includes the inflection of text on voiced consonants for expressive purposes, the singer may have momentarily produced a voiced sound on a consonant at the onset or offset of a word at a much lower [F.sub.0] and SPL level than she might sing on a vowel. Finally, there is the possibility that pitch extraction errors occurred with the software used for this plot of the song performance. The musical score indicates a range of [D.sub.4] (293.66 Hz) to [A[flat].sub.5] (830.61 Hz). The TF32 [F.sub.0] extraction from the SLM signal indicated a song performance range of 50.17 Hz-862.76 Hz, while the Kay software extraction of [F.sub.0] from same SLM signal ranged from 192.64 Hz-936.06 Hz (although only eight values occurred above 860 Hz).

Estimated versus Measured Time and Cycle Doses and Short Term Recovery

Table 3 compares the estimated values for voicing time dose, cycle dose, and short term recovery from the tessituragram with the measured values of the song's performance, and also provides a comparison of values captured from the Iasus throat microphone with those captured through the Quest SLM. While the time and cycle dose estimates were relatively accurate, the recovery estimates were substantially less than the actual values. This was perhaps due in large part to short unvoiced segments which were not indicated by the musical notation used to derive the recovery estimate. An examination of the text of the song revealed at least 60 instances where unvoicing would occur (e.g., "und vor Lust hinein zu schauen") for German diction purposes outside of the indicated musical rests. Another possibility for some of the short unvoiced segments is [F.sub.0] extraction error; as a note, the TF32 software was used for all of the dose measurements.

Table 4 displays the difference in values for time dose, cycle dose, and recovery between the Iasus throat and Quest SLM microphones. As is shown, the voicing measurements are all greater for the throat microphone than for the airborne signal captured by the SLM, while this pattern is reversed for the recovery time measurements. One possible explanation for these differences is that the throat microphone's direct contact with the singer's neck tissue might have provided greater sensitivity to low intensity phonation at voicing onset and offset than the SLM microphone; thus, voicing measured by the throat microphone signal might at times momentarily precede voicing onset measured by the airborne microphone, and continue momentarily after the airborne signal already indicated voicing offset.


As was stated in the introduction, repertoire selection for singers is a subtle, multifaceted process. The methods used in this pilot study, particularly performing the tessituragram estimates, were labor intensive, and would require additional refinement before they could be practical for voice instructors and performers to use. Perhaps an estimation method could be developed for automatically extracting pitch and rest occurrences and durations using information from digital files of musical scores (e.g., Finale, Sibelius, MIDI); if so, the tessituragram procedure could be accomplished relatively simply.

Commercial dosimeters (e.g., the Kay Ambulatory Phonation Monitor) were not used for this study, either, as the cost of such a device is still prohibitive for most voice teachers (including the author and his academic institution). Use of a dosimeter could streamline the dose extraction process substantially. More cost- and time-effective in the near term than dosimeters might be the system devised by Lamarche, Ternstrom, and Hertegard, which combined a commercial VRP program with a response button for the singer to use during difficult passages. A singing teacher could use information gathered by this process to more precisely select pieces for the student with the optimal ratio of technical challenges to technical strengths.

Other factors than physiological [F.sub.0] and SPL range and dose measurements are used by voice teachers when deciding whether a singer should perform a particular piece. A number of these could prove challenging to quantify:

(1)The distribution of ascending versus descending direction phrases occurring in a piece, and the relative energy cost to the singer for each type of phrase. Repeated ascending phrases often invite a more "full" or "heavy" production, physiologically correlating perhaps with more TA activity. While the spectrogram and the EGG could be used to indirectly monitor intrinsic muscle balance in real time (via changes in the strength of higher partials and the magnitude of the closed quotient, respectively), Titze's tessituragram method, which is helpful in assessing the piece before the singer attempts the vocal selection, does not yet include a means of weighting the phrase directions found in a piece.

(2) The difficulty of the musical approach to the extreme passages (highest or lowest). Extreme notes in a song or aria can be approached scalewise by step, by leap to the high/low notes, or by initiating directly on the extreme notes. In the author's experience, the preferred approach varies from singer to singer. Many seem to prefer a leap of less than an octave (typically a 3rd, 4th, 5th, or 6th) when approaching a high note, while a stepwise descent is often preferred as a lead-in to a low note. Again, weighting or scaling factors have not yet been developed to account for this in tessituragrams. Examining the music score with the singer's strengths or preferences in mind would likely be preferable and more practical to any type of quantifying process.

(3) The vowels on the highest/lowest pitches. Given formant locations and tuning strategies employed by successful singers, preferred vowels for extreme notes are very much a gender specific issue. A brief written comment listing the vowels on highest and lowest notes provided as a supplement to a tessituragram would enhance the tessituragram's value to teachers and singers more than any numeric value or graph. A similar statement could be written regarding the consonants associated with higher/lower pitches and the manner in which the text is set in extreme passages. For example, due to the acoustic reasons mentioned above, most females prefer a melismatic rather than a syllabic setting of text on high pitches, in order to avoid having to close their mouths (lowering the first vowel formant) to articulate consonants.

(4) The presence of florid high or low passages versus sustained high or low passages. While a tessituragram using Titze's method indicates where the preponderance of the singing occurs, it does not necessarily indicate what kind of singing is occurring. It is possible that a florid high piece by Rossini and a soaring, legato high piece by Strauss might have similar tessituragrams in terms of cumulative time and cycle doses for each extreme note. More pieces need to be assessed to test this possibility, as the type of singer who excels at Rossini is seldom a Strauss specialist, too.

(5) The complexity of the musical texture/richness of the accompaniment. This factor is linked to the intensity level required to sing key phrases. Examination of the score is required.

A topic only briefly mentioned in section two of the methods is the dynamic level. If the composer's dynamics were provided, an estimate of the time dose and cycle dose sung at each dynamic level could be calculated for a piece. Relating estimates derived from musical dynamic markings to a vocal fold tissue movement distance dose would be more problematic, however. Commercial dosimeters typically used for speaking voice research and clinical applications employ normative data for vibration amplitude when they calculate a distance dose based upon average SPL, but the vocal intensity of a singer depends upon many factors, one of which is the expert adjustment of the singer's vocal tract with changing vowels and pitches. Any distance dose estimates or measurements on singers derived from SPL levels must take into consideration the possibility for substantial errors.

Finally, although it is outside the scope of the physiologically oriented discussion in this article, singing teachers and their students constantly have to consider less quantifiable elements of vocal performance when they select repertoire, such as the singer's facility or lack thereof in a specific language, the suitability of the sung text for the singer's gender, age and personality, the career aspirations of the singer, and the singer's overall musicianship..


The VRP, tessituragrams, and dosimetry deserve recognition as helpful tools for repertoire selection for teachers of singing and their students. Widespread use of these tools by teachers will depend upon further hardware and software advancements, upon more singing teachers becoming familiar with objective measurements of the singing voice, and upon singing teachers and students learning how to better interpret objective data for guiding actions.


(1.) John Nix, "Criteria for Selecting Repertoire," Journal of Singing 58, no. 3 (January/February 2002): 217-221; John Nix, "Vocology and the Selection of Choral Repertoire," Australian Voice 13 (2008): 36-42; Janette Ralston, "Development of an Instrument to Grade the Difficulty of Vocal Solo Repertoire," Journal of Research in Music Education 47, no. 2 (Summer 1999): 163-173.

(2.) Richard Boldrey, Guide to Operatic Roles and Arias (Redmond, WA: Pst, 1994); Joan Frey Boytim (ed.), Solo Vocal Repertoire for Young Singers: An Annotated Bibliography (Jacksonville: National Association of Teachers of Singing, 1982); Berton Coffin and Werner Singer, Singer's Repertoire, 2nd ed. (New York: Scarecrow Press, 1960-62); Barbara Doscher, From Studio to Stage: Repertoire for the Voice (Lanham, MD: Scarecrow, 2002); Shirlee Emmons and Stanley Sonntag, The Art of the Song Recital (New York: Schirmer Books, 1979); Noni Espina, Repertoire for the Solo Voice: a Fully Annotated Guide to Works for the Solo Voice Published in Modern Editions and Covering Material from the 13th Century to the Present (Metuchen, NJ: Scarecrow, 1977); Thomas Goleeke, Literature for Voice: An Index of Songs in Collections and Source Book for Teachers of Singing (Lanham, MD: Scarecrow, 2002).

(3.) Robert Glaubitz, (accessed February 8, 2013); Stephen Lancaster, com/site/stephenlancastervoicestudio/vocal-repertoiredatabase (accessed February 8, 2013).

(4.) Kate Emerich, Ingo Titze, Jan Svec, Peter Popolo, and Gary Logan, "Vocal Range and Intensity in Actors: A Studio Versus Stage Comparison," Journal of Voice 19, no. 1 (March 2005): 78-83.

(5.) Anick Lamarche, Sten Ternstrom, and Stellan Hertegard. "Not Just Sound: Supplementing the Voice Range Profile with the Singer's Own Perceptions of Vocal Challenges," Logopedics Phoniatrics Vocology 34, no. 1 (February 2009): 3-10.

(6.) Anick Lamarche, Sten Ternstrom, and Peter Pabon, "The Singer's Voice Range Profile: Female Professional Opera Soloists," Journal of Voice 24, no. 4 (July 2010): 410-426.

(7.) Christian Herbst, Elke Duus, Harald Jers, and Jan Svec, "Quantitative Voice Class Assessment of Amateur Choir Singers: A Pilot Investigation," International Journal of Research in Choral Singin[G.sub.4], no. 1 (Fall 2012): 47-59.

(8.) Thomas Carroll, John Nix, Eric Hunter, Kate Emerich, Ingo Titze, and Mona Abaza, "Objective Measurement of Vocal Fatigue in Classical Singers: A Vocal Dosimetry Pilot Study," Otolaryngology Head and Neck Surgery 135, no. 4 (October 2006): 595-602.

(9.) Matthew Schloneger, "Graduate Student Voice Use and Vocal Efficiency in an Opera Rehearsal Week: A Case Study," Journal of Voice 25, no. 6 (November 2011): e265-e273.

(10.) Christopher Gaskill and Jennifer Cowgill, "Vocal Dose Measurements with Undergraduate Music Majors," Kay/Pentax Ambulatory Phonation Monitor: Applications for Voice and Speech (October 2009): 1-3.

(11.) Christopher Gaskill, Jennifer Cowgill, and Shenendoah Many, "Comparing the Vocal Dose of University Students from Vocal Performance, Music Education, and Music Theater," Journal of Singing70, no. 1 (September/October 2013): 11-20.

(12.) Christopher Gaskill, Jennifer Cowgill, and Sara Tinter, "Vocal Dosimetry: A Graduate Level Pedagogy Course Experience," Journal of Singing 69, no. 5 (May/June 2013): 543-555.

(13. Stefan Thurmer, "The Tessiturogram," Journal of Voice 2, no. 4 (1988): 327-329.

(14. Ingo Titze, "Quantifying Tessitura in a Song," Journal of Singing 65, no. 1 (September 2008): 59-61.

(15.) Kevin Hanrahan, "Use of the Voice Range Profile in Assigning Repertoire: An Evaluation" (Paper presented at the annual meeting of the ISME World Conference and Commission Seminars, Beijing, China, August 1, 2010); available at: index.html (accessed January 21, 2013).

(16.) Nico Paolo Paolillo and Frano Fussi, "The Vocal Score Profile/Vocal Range Profile Rate and APM in Artistic Voice Evaluation: Application Tested on Opera and Musical Singers; An Evaluation of Voice Suitability and Vocal Fatigue," in Claudia Manfredi, ed., Proceedings of the Seventh International Workshop on Models and Analysis of Vocal Emissions for Biomedical Applications (Firenze, Italy, August 25-27, 2011), 85-92.

(17.) Ingo Titze, Principles of Voice Production (Englewood Cliffs, NJ: Prentice Hall, 1994), 219-220.

John Nix, tenor, is Associate Professor of Voice and Vocal Pedagogy at the University of Texas at San Antonio, coordinator of the Vocal Area for the 2011-2014 academic years, and founding director of the UTSA Vocal Arts Laboratory. Previously he was on the staff of the National Center for Voice and Speech in Denver, where he worked with Ingo Titze. Mr. Nix has also served on the music faculties of The University of Colorado at Denver and Eastern New Mexico University. He holds degrees in Arts Administration from Florida State University, in Vocal Performance from the University of Georgia and the University of Colorado at Boulder, and Certification in Vocology from the University of Iowa. At Colorado, he studied voice and pedagogy with the late Barbara Doscher and the Alexander Technique with James Brody. His current and former students include a member of the Mormon Tabernacle Choir, two Santa Fe Opera apprentices, members of the Army Soldiers' Chorus, a second place winner in the National Federation of Music Clubs competition, a two-time finalist in the American Traditions competition, and faculty members at universities in Montana, Texas, Wyoming, and New York. UTSA students of his have gone on to win graduate fellowships to major universities. His work has been funded by The San Antonio Area Foundation, The Grammy Foundation, UT-San Antonio, and two R-13 grants from NIH. Mr. Nix was the 2006 winner of the NATS/Voice Foundation Van Lawrence Award. His published articles have appeared in The NATS Journal, The New York Opera Newsletter, Otolaryngology-Head and Neck Surgery, Journal of Voice, Journal of Singing, International Journal of Research in Choral Singing, VocalEase, Australian Voice, and Opera Journal. Mr. Nix is editor and annotator of From Studio to Stage: Repertoire for the Voice, compiled by Barbara Doscher (Scarecrow, 2002), is Vocal Music section editor for the Oxford Handbook of Music Education (2012), and a general editor and author for The Oxford Handbook of Singing (to be published in 2014).

[Associate Editor's note: In the September 2008 edition of "Voice Research and Technology," I introduced a topic called "Quantifying Tessitura in a Song." It was more of a teaser than a complete work. My good friend John Nix completed a real study, including dosimetry on a singer who performed a Mozart composition (different from "Il mio tesoro intanto," which I used for my test case)--Ingo R. Titze.]

National Association of Teachers of Singing


Membership in the National Association of Teachers of Sing ing is open to any citizen of any country whose professional train ing and experience qualify ing him or her as a teacher of singing.

Associate Membership is available for voice teachers and advanced students who have not as yet completed the requirements for full membership.

Affiliate Membership is open to persons or groups that are interested in vocal pursuits but are not actually involved in the teach ing of singing, such as speech therapists, laryn gologists, schools, publishers, and music stores.


Membership includes a sub scription to the Journal of Singing, the official journal of NATS, and to Inter Nos, the NATS Newsletter.

Applications for membership may be completed online at The website contains detailed information about the qualifications for membership and the Code of Ethics of the National Association of Teachers of Singing. Visit the Membership section of to learn more.

TABLE 1. Time and cycle dose estimation.

  Estimated time and cycle doses for a female singer for "An Chloe"

Musical              F0       No. of    Duration in sec   Cycle dose
pitch               (Hz)       beats     at half = 60

[D.sub.4]           293.66      1         0.5               146.83
[E[flat].sub.4]     311.13     19.5       9.75             3033.5175
[E.sub.4]           329.63      0         0                   0
[F.sub.4]           349.23      9.25      4.625            1615.18875
[G[flat].sub.5]     370         0         0                   0
[G.sub.4]           392        24        12                4704
[A[flat].sub.5]     415.3      17.25      8.625            3581.9625
[A.sub.4]           440        10.5       5.25             2310
[B[flat].sub.5]     466.16     50        25                1654
[B.sub.4]           493.88      2         1                 493.88
[C.sub.5]           523.25     12.5       6.25             3270.3125
[D[flat].sub.5]     554.37      2.5       1.25              692.9625
[D.sub.5]           587.33     14         7                4111.31
[E[flat].sub.5]     622.25     26        13                8089.25
[E.sub.5]           659.26      0         0                   0
[F.sub.5]           698.46      6         3                2095.38
[G[flat].sub.5]     740         0         0                   0
[G.sub.5]           784         1.5       0.75              588
[A[flat]sub.5]      830.61      1         0.5               415.305
Time Dose                      98.5
Cycle Dose                                                46801.9

TABLE 2. Estimated short term recovery.

             Estimated short term recovery for "An Chloe"
    Fermati rendered as double the written value of the note (per
                   traditional practice)
Rest              Duration               Number of        Total recovery
                  (at half               occurrences      in seconds
                  note = 60)
                  in seconds

eighth               0.25                     5                1.25
quarter              0.5                     17                8.5
dotted quarter
or equivalent        0.75                     0                0
half or
equivalent           1                        5                5
dotted half or
equivalent           1.5                      1                1.5
whole or
equivalent           2                        1                2
5 beats              2.5                      1                2.5

TABLE 3. Estimated versus actual dose measurements and short term

  Estimated versus actual time and cycle doses from throat mic and SLM

Dose Type       Estimated       Actual dose            Actual dose(SLM)
                @ half          (throat microphone)    and % difference
                note=60          and % difference      from estimated
                                 from estimated         value

Time dose    98.5seconds       107.6seconds(+9.2%)    105.4seconds(+7%)
Cycle dose   46801.9 cycles    50296.4 cycles         49100.3 cycles
                               (+7.5%)                (+4.9%)
percentage                         75.35%                   74.09%

              Estimated versus actual short term recovery

               Estimated     Actual recovery       Actual recovery(SLM)
               @ half        (throat microphone)   and % difference
                note=60       and % difference      from estimated
                              from estimated        value

                20.75 sec     35.19 sec (+69.59%)   36.86 sec (+77.64%)

Recovery time      Estimated       Actual number       Actual number
in seconds         number of       of occurrences      of occurrences
                   occurrences   (throat microphone)   (SLM)

0.001-0.05             0              189                180
0.05-0.24              0               70                 81
0.25-0.49              5               22                 20
0.5-0.75              17                8                  7
0.75-0.99              0                1                  4
1.0-1.49               5                5                  3
1.5-1.99               1                2                  4
2.0-2.49               1                0                  0
2.5-3.0                1                0                  0

TABLE 4. Differences in measured doses and recovery values between the
throat microphone and the SLM.

                   Phonation                                Short-Term
                   Time Dose                Average         Recovery
Measurement        (seconds)   Cycle Dose  [F.sub.0](Hz)    (seconds)

Throat microphone  107.6       50296.4     462.02           35.19
Sound level meter
microphone         105.4       49100.3     459.79           36.86
Difference           2.2        1196.1       2.23           (1.67)
COPYRIGHT 2014 National Association of Teachers of Singing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Nix, John
Publication:Journal of Singing
Geographic Code:1USA
Date:May 1, 2014
Previous Article:Healthcare teamwork, insights, and challenges.
Next Article:Singing in Danish: a guide to diction.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters