In comparison, non-Suzuki students had a far better grasp of note names and all non-Suzuki students could read music sufficiently to read pieces requested of them. All students were assigned a lesson for each feedback style during one their four lessons. Lessons were randomly ordered within the constraint that they were equally distributed. Lessons were designed to maintain a balance between tasks to keep students interested and learning while ensuring sufficient repetition of tasks to enable effective comparisons.
Each lesson, consistent across feedback styles, consisted of the consecutive three parts in Table 1. Structure of study lessons split into sections with task, repertoire, key, and time spent within a lesson section. In this table, key refers only to musical segments included in overall intonation results for that part of the lesson section. Two of the three included sections consisted of fixed repertoire in order to be able to compare results across lessons and participants more directly. Scales in participants' lessons alternated between G Major and A Major.
All but three participants were familiar with one or both pieces prior to the study. Though students played the full piece, collated results of a lesson section used only the music's A section so that results for a section are always limited to a single key. Tasks in unstructured time were decided based on what the student needed help with, which may or may not pertain directly to the study. The amount of time and number of repetitions within each section depended on the student and the lesson length. Students unfamiliar with the chosen common repertoire piece might spend most of the lesson working on it and largely skip unstructured time, while more competent students might speed through required tasks and spend most of the lesson in unstructured time.
Similarly, if time was short, sections were moved through faster. In a typical lesson, students would play three scales, followed by three arpeggios. As arpeggios were less familiar, these were often repeated additional times. The common repertoire piece was frequently repeated unless a student displayed a high level of proficiency playing it. Within this study we used equal-temperament tuning for detecting error and tuning aural feedback. Expert musicians who do not play instruments with fixed intonation are well-known to use different tuning systems such as Pythagorean, 6th comma meantone, and expressive intonation Loosen, ; Kopiez, ; Leukel and Stoffer, ; Duffin, ; Johnson, In many of these tuning systems, tuning becomes contextual; the tuning of a note may depend on the key or its melodic role.
In contrast, equal-temperament tunings are consistent across all scenarios. We do not expect beginning violinists to be familiar with the subtleties of different temperament systems, and given the ubiquity of equal temperament on the piano and in wider musical culture Duffin, , 19 , we felt the use of equal-temperament was a reasonable compromise for defining what was correct within the study even if some deviation from equal temperament which would be marked as an error in this study might be positively perceived by an expert listener.
Based on pre-study trials, when the performance task featured pre-dominantly single-key music scales, arpeggios, and common repertoire , we corrected pitches to the nearest equal-temperament note in the relevant key.
- Part 3 - Scales and Arpeggios.
- Step 1: What You Need to Know First.
- Contro la meritocrazia (Paceinsieme) (Italian Edition).
- Subscribe Our Newsletter!
It is common for beginners to play an incorrect accidental, and snapping , pulling a pitch to the nearest target note, to key meant that the pitch corrected audio should correct to the proper note. As one beginner in stated during pre-trials in response to snapping chromatically:.
I didn't know if I was sharp or flat because it was correcting me to the wrong note. And as a beginner, …I knew it was wrong, but I didn't know why. We logged two timestamped data streams from the augmented violin: one for sensor readings and one for pitch estimation. Sensor readings of finger contact location were logged every Direct audio of the performance and aural guide along with video were recorded for every lesson.
Results for pitch were calculated based on the logged equal-temperament linearized estimates of pitch performed 8. When calculating intonation accuracy, we removed any pitch estimates logged during periods of low volume, presumably rests of musical pauses, or that were clearly incorrect namely those above or below the possible played range. As much of what we are interested in is the intonation correction process, we calculated continuous frame-by-frame intonation error for the duration of a note.
Intonation error was determined as the estimated difference in cents between the pitch played and the nearest chromatic pitch. For example, if A is defined as our zero reference point, each equal-tempered semitone will have an integer pitch value, such that the C above Hz has a value of 3. Thus, a linearized pitch estimate of 3. There was one significant issue with the augmented violin implementation that arose and was addressed during the study.
Freebie: Technical Requirements Charts for RCM's Piano Syllabus - Color In My Piano
Despite rigorous pre-testing with adults, as lessons with the half-sized violin progressed, it became apparent that it was challenging for non-adults to sufficiently press down the string to trigger the fingerboard sensor when playing the first finger. Due to how our low-latency pitch detection algorithm was optimized, when the first finger was not pressed down adequately, it resulted in a pitch estimate up to 70 cents flat.
Once detected, the issue was fixed, but it persisted for exactly half of the lessons. The impact varied by participant and by lesson. Numerical results in the study are based on the revised algorithm but participant experiences may have been affected by which algorithm was used during their lessons. Questions were intended to assess whether students benefitted from and enjoyed feedback and to facilitate conversation.
Feedback was also discussed within the lesson as teaching situations arose. Planned questions asked after the completion of each lesson were:. The last lesson also included a final interview on the overall experience. Feiring conducted final interviews with his students P1, P4, P7 while at the Suzuki Hub, for practical reasons final interviews were primarily asked by the author.
Suzuki Hub director Kate Conway re-interviewed three participants and one student was re-interviewed by her mother. The four overall study questions are listed in List 2. Lessons and interviews were recorded on video and annotated with transcribed dialog and information about how long each section in Table 1 lasted, whether the student was clearly looking at or away from any visual feedback, and whether the teacher was playing with the student.
Qualitative results were derived using thematic analysis Braun and Clarke, Having extracted information on which feedback method was a student's favorite in the context of their lesson experiences and which they thought they might prefer during independent practice, all annotations of conversations and events were reviewed for common themes by the primary author. We conducted two levels of review for themes; the first was themes within or specific to a given feedback method, and the second was themes shared across feedback methods. Comments clearly about a specific single feedback style, such as aural only, or visual only, made during a lesson using combined feedback were considered relevant to the particular feedback method, rather than only in the context of mixed modalities.
Once potential themes were collected, they were reviewed for clarity and significance.
Clarity was required not just whether the particular theme was well-defined, but whether the meaning of the intention of the student's comments contributing to the theme were sufficiently clear and independently generated. If a question was deemed too leading or pressured, the student's response was thrown out.
Significance was given to themes either repeated by a large number of students, or generated by a strong response by a small number of students. Themes directly from participant feedback form the structure of section 3. Fourty-nine lessons were completed with 47 lessons included in overall results. Two lessons were repeated due to issues with study execution one is discussed in section 4. Numerical analysis of intonation accuracy was segmented per lesson section into scales and arpeggios, an introductory common repertoire piece, and unstructured time Table 1.
For each participant, we derived the mean absolute intonation error per section with repeated scales, arpeggios, or pieces grouped together within each section. The overall error for a participant's lesson was taken as the mean across all sections. The mean errors from each participant were then used to calculate the mean absolute intonation error across all participants for all sections, as shown in Table 2.
Scales and common repertoire are measured against the key of the section and have a maximum error of cents, while error during the unstructured time is measured against a chromatic scale with a maximum of 50 cents. Error was measured against key except during Unstructured Time which was measured chromatically. Lowest Err is the number of participants for whom the given feedback method resulted in their lowest MAIE for that section. Numbers for preferred lesson or practice represents the number of participants selecting the given feedback method as their favorite.
Lowest MAIE per section is in bold. Our experimental data tracking performed pitch did not provide evidence to support our hypothesis that overall, a beginner violin student will perform with better intonation provided with aural, visual, or both forms of feedback. Considering the overall mean absolute intonation error for each of the four feedback cases, the maximum difference between cases was only 1.
With such high variance, we would need a much larger sample size for conclusively determining whether our hypothesis was true or false. Figure 3 depicts performance results on an individual basis, highlighting the divergent response of participants to different feedback mechanisms. The overall means for each feedback condition masked a significant variation in performance both between participants and within an individual's set of lessons Figure 3. More meaningful than statistical analysis in our case is to look at trends amongst participants. For instance, P8, performed dramatically worse in the lessons with audio feedback than without.
In fact, P8 accounted for a substantial amount of the overall difference in performance between aural feedback and no feedback cases in Table 2. Removing P8, the difference drops to 0. Though no certain conclusions can be drawn, it is reasonable to hypothesize that for P8, aural feedback had a negative impact on intonation performance. Overall mean absolute intonation error per lesson for different feedback methods for each participant.
As in section 2. Individual responses are given in Table 3. Participant stated preferences for favorite lesson type and what feedback method s they speculated they would most like during individual practice. L means participant responded the given feedback method was their favorite lesson style, while P means given feedback method was their preferred option for potential use during practice.
Fifty-eight percent of participants reported aural feedback as their favorite lesson type Figure 4. Only one student reported preferring no feedback with two opting for visual and two for combined feedback. While the question about favorite lesson type seemed to lead to single modality answers, when changing the question to what a student thought would be most helpful during practice, two-thirds of participants stated a preference for having aural and visual feedback.
Two participants reported they would probably only use visual feedback during practice, with one P1 explicitly saying he did not like the aural feedback as, though sometimes he found it useful, it was also distracting and he did not trust it. Two participants reported they were not interested in having any intonation feedback method while practicing.
Reported preferences of the 12 participants for favorite lesson within feedback types and what feedback option s would be most helpful during practice. Combining responses from both questions about feedback preference in Figure 4 , aural feedback was included in With no feedback only chosen in We split review of participant feedback and responses into themes.
The Manual of Scales, Broken Chords and Arpeggios ( for Piano )
After a brief discussion of bias and we look at some of the primary themes that arose: utility, attention, familiarity, sensory overload, immediacy, and trust and authority. Prior to discussing user feedback, as suggested in section 2. For instance, answers may favor more recent lesson types or be influenced by interviewer comments. There is evidence of this within the study such as numerous contradictory statements, typically between different lessons. For instance three participants stated at different times they liked either aural or visual feedback better than the other only to reverse their preference 9.
With those caveats, we assume that students are truthfully reporting their experiences. In interviews, we always endeavored to ask students to justify and explain simple answers, not only in an effort to gain wider insight, but also to make sure they could provide reasonable rationale for their answers. Without rationale, we did not weigh yes or no answers strongly.
Alternatively, we found examples of students P1, P8 taking a contrarian stance, only reluctantly admitting to use of feedback. While aural feedback resulted in the worst intonation accuracy for 5 out of 12 participants P3, P6, P8, P10, P12 , it was by far the most popular lesson type 7 out of In response to questions from List 2 and question derived from List 1, five participants P2, P3, P5, P11, P12 , asserted that, as intended, aural feedback highlighted error and provided practical feedback for correcting.
Because it sort of tells me the right note, like the equivalent to what I would be playing. Except I'm doing it for a shift so if my shift is out of tune it tells me the note. Which helps a lot. If I notice that I'm playing out of tune, it tells me whether I'm flat or sharp. Participants also generally commented positively about visual feedback 10 out of 12 as it also informed participants whether they were in tune.
The graphic design was praised as easy to interpret: the colored block below the in-tune line meant raise the note, and above the line meant lower it. Even if it is just slightly and you can't hear it and it is good for getting it really in tune to perfect a piece. Participants overwhelmingly selected aural and visual feedback as their hypothetical preference for practice 8 of When aural feedback was confusing, having visual feedback allowed the participant to clarify what was happening, and having aural feedback was useful when playing with a score or unable to devote sufficient visual attention.
As P11 said,. A major problem with visual feedback was that it required active visual attention. In order to use visual feedback, the student must be able to look at it. Half the participants P1, P2, P4, P5, P11, P12 confirmed that it could be challenging to use visual feedback when visual attention was required elsewhere. Gaze annotated during lessons, a sampling of user visual attention, suggested during lessons using visual feedback, the graphics did not regularly receive attention. Though indicative of use, our simplified measure of gaze is not necessarily fully inclusive; for instance, P4 remarked he watched the feedback out of the corner of his eye while focusing on the score.
Participants most commonly looked at visual feedback during scales and arpeggios which were taught aurally and did not require use of a score. As a result of issues with visual attention, the consensus amongst the two adult participants, P2 and P11, was that visual feedback was only an effective modality of feedback once the notes of a piece had been learned. When starting a new piece, wearing the headphones was far more useful. In contrast, four students expressed that they appreciated that aural feedback was essentially a passive yet present form of feedback P2, P4, P5, P11 : using it did not require an intentional direction of attention.
Still, part of making aural feedback effective and ensuring attention is appropriate volume. Too soft and the participant can not use it effectively, yet too loud and it becomes painful and distracting. For instance, P2 pointed out in a lesson that he could not hear the guide sufficiently and requested it to be louder, while P8 had the volume distressingly loud for most of lesson.
Sensory overload was an issue for both aural and visual feedback. In cases where the student expected the teacher to play with them, primarily in lessons with Suzuki students, the author tried to play quietly so as not to interfere with or overshadow the aural feedback, however it may have still been problematic. Additionally, while aural feedback audio was generally of good quality and free from artifacts, it still sounded different.
Besides tonal differences, not only were there times when it burbled on low strings as a result of pitch tracking error, but ambient sounds from the room were amplified and potentially pitch corrected as well. Participants were told they were free to remove headphones if they wanted. Two students requested to remove the headphones. P4, who liked the aural feedback but would frequently remove his headphones when not playing, explained there was background noise which he found unpleasant and hard on the ears.
In future, efforts need to be made to eliminate or reduce background noise in aural feedback. Only P8, requested to take them off fully: once as it had been too loud and the other time he stated he disliked the aural guide, the headphones were bulky and the cable annoying. In both cases he completed all but the unstructured time wearing headphones. P8 was also the one participant whose intonation accuracy was dramatically worse in lessons using aural feedback see Figure 3. Similarly, visual feedback also suffered from sensory overload.
Half the students expressed that visuals were distracting due to feedback changing too rapidly for users to process and flicker due to estimation error between notes. For instance P11 remarked,. It's interesting. Obviously the color and the direction of the bar does help, but in a playing situation, it's way too much and it's flickery and it makes me just really confused, whereas if i just focus on what I'm hearing it's much easier to play it in tune…it was all red bars everywhere and it was going too fast to correct it whereas if I'd of been able to hear it I could have corrected it.
Another issue with visual feedback, noticed by P2, is that his reaction time to visual feedback was slower. Above some tempos, aural feedback will also be unusable, but aural feedback has the advantage that we can process auditory stimuli faster. In comparison, it takes 50 ms for humans to trandsuct visual stimuli in normal light. Hanson et al.
Technologically, audio delay was also less than graphics. The audio latency of the pitch-correction software was Further, intonation data required a low-pass filter to be stable enough for human perception. Combined feedback will suffer from the same sensory overload problems that both aural and visual feedbacks have independently. The combination will not fix flicker in visual feedback, audio noise in the aural feedback or potential overstimulation, but having both does give the participant the option to switch focus if desired.
In fact, there was some evidence that trying to use both feedback methods simultaneously was itself overwhelming P2, P Familiarity turned out to play a significant issue for both feedback methods, but in different ways. Despite largely positive response to aural feedback, both adult participants mentioned it was necessary to habituate oneself to having an additional audio source and manage potential sensory overload.
Strong evidence that aural feedback takes some acclimation comes from separating lessons further into component parts, splitting intonation performance between scales and arpeggios Table 4. As stated in section 3. Reduced version of Table 2 giving the MAIE of different sections with scales and arpeggios separated. Improvement in intonation error throughout the lessons without aural compared to reduced improvement in cases with aural feedback suggests it may take time for a user to become accustomed to aural feedback.
In both methods using aural feedback, intonation accuracy during scales was more than 2. This gap is the largest difference between group means seen in this entire study. Intonation in aural inclusive methods improved in comparison to non-aural methods during arpeggios, and by the common repertoire section, intonation in aural inclusive modalities improved to the extent that they were largely on par with non-aural modalities.
Improvement between scales and common repertoire for aural feedback inclusive methods was nearly double the improvement seen in both non-aural cases. Some of this difference may be that, while visual feedback uses a sensory modality which is not strictly required for performance, aural feedback alters a key, already in use mode of feedback. Participants must become accustomed to hearing their violin in only one ear while hearing a compressed pitch corrected version which does not necessarily match their playing in the other.
While participants were used to playing with teachers and hearing two people playing, participants reported aural feedback was quite different P8, P It was much easier to hear and see. Because of the compression, I could tell which one I was aiming for really clearly. When I had got into it, I was getting to grips with the feedback and was getting much better at correcting my pitch.
Familiarity was a challenge with visual feedback for a different reason: unfamiliarity with note names. Participants found the basic bar graphic described and pictured in Figure 2 easy to use, but note names were of limited use. Only 3 study participants, P4, P6, and P11 showed competence naming notes in a score. All other students referred to notes by their first position fingering. Even more experienced players displayed poor knowledge of note names.
This is an area we expected Suzuki students to be weaker and indeed, despite making up two-thirds of participants, only one of the students recognizing note names was a Suzuki student. The problem with note unfamiliarity is that most students could therefore not determine whether they were playing the correct note or not. This expectation usually proved untrue thus reducing the usefulness for correcting highly out of tune notes. However two players pointed out displaying previously unfamiliar note names may provide a learning opportunity. I will know the notes; how to say them in the letters.
Participant comments suggested that one reason for liking visual feedback over aural feedback was that it was more explicit. Three users P2, P8, P11 expressed that it was sometimes difficult to decide which way to fix a note using aural feedback whereas with visual feedback it was easy to interpret as it showed in which direction to move their finger. P11 stated:. It's enough information for me to correct it…like I knew I was out of tune when you were just teaching me without anything, without any aids, but I didn't know how to correct it, whereas this is really clear how to correct it…[With] the headphones, I could hear that I was out of tune and that made it quite stressful, where as [with visuals], there is only one thing I'm listening to, it is just my note which is easier to process and really clear …whether I am too high, too low.
P5 also suggested that aural feedback could be stressful as one might hear they were wrong but did not always know the best way to correct it. Trust is crucial for intonation feedback tools to be useful. If a practice aid is frequently wrong, it loses its value. This was true even as it struggled to correctly identify the pitch of the three lowest notes on the violin. Asking P6, who has perfect pitch, about visual feedback at the bottom of his scale:. Trust and authority were necessary to encourage participants to use feedback, but they had the drawback that participants were less likely to report issues during the experiment.
For instance, with aural feedback, there were multiple times the author initially failed to correctly set the key for aural feedback, yet only two participants P4, P11 ever pointed the mistake out. Similarly, if headphone volume was uncomfortably loud, rather than speaking up the participant would often tolerate it through out most of the lesson or until the author sensed something was amiss P8, P9. Despite trust being high, two participants P1, P8 expressed they thought the system was sometimes wrong.
Neither of them said they liked the aural feedback, with P1 preferring no feedback at all. Active use of aural feedback is difficult to assess outside of self-reporting since listening is not externally visible.
Piano - Scales and Arpeggios-royal Schools of Music
As an observer, it can be difficult to infer whether an improvement in intonation accuracy is due to the enhanced feedback tools, the student's self-assessment based on their internal memory of the task, teacher instruction, or another contextual factor. Especially as out quantitative data is inconclusive as to whether aural feedback has any effect on intonation error and therefore any indication whether it is being used or not, it is valuable to confirm its active use outside of self-reporting. Being a manual intervation, there were occasional instances in the study where the key for a task was temporarily set incorrectly.
Repeated or uncharacteristic intonation error triggered the primary researcher to ask whether the key was set correctly. Popular Features. New Releases. Contains all scales and arpeggios for the revised syllabus from , with bowing patterns and suggested fingering, and a helpful introduction including advice on preparing for the exam. Other books in this series. Add to basket. The syllabus now takes the player through the whole range of the double bass, from a one octave E major scale starting on the lowest open string for Grade 2 to E major in 3 octaves at grade 8.
The rhythmic changes for scales published in such as way that it is still possible to choose the 'old' style without bamboozling either student or teacher. Scales are printed twice, firstly in the new version as even quavers and then on the next page in the old style. Visually, the finger look les cluttered on the page as, unlike previous editions, fingerings are not marked on every note.
Recorder Blocke-flote? Alto Saxophone. Bass Saxophone. Tuned Percussion. General Percussion. Ethnic percussion. Folk instruments. Tin Whistle. Concert Band.
- Sign up to receive news & offers;
- A Forbidden Love (Christas Journey Book 1).
- Christmas With Southern Living: Savor * Entertain * Decorate * Share?
- Do you neglect scale and arpeggio practice because you find it tedious?.
New releases. Composer A-Z. Arranger A-Z. Publisher A-Z. Flexible band. Brass Band. Fanfare Band. Marching Band. Jazz Ensemble. Flexible ensemble. Wind ensemble. Woodwind ensemble. Brass ensemble. Percussion ensemble. Accordion ensemble. Other ensemble. Vocal Scores. Full Score.