Resolving Conflicting Linguistic and Musical Cues in the Perception of Metric Accentuation in Song

Jieun Oh Center for Computer Research in Music and Acoustics Stanford University

Analysis on the Korean Translation of Happy Birthday to You

On the interplay of accentuation in language and in music

From Dell and Halle (2005)

lyrics (text)

2

vs. melody (tune)

Typical resolution of conflict: �Tune > Text

3

  Ross, J. (2003) Rhythmic information for Estonian

  Mang, E. (2007) Pitch information for Chinese

Atypical resolution of conflict:�Text > Tune

4

  Korean translation of “Happy Birthday”

(1) Original lyrics

(2) mismatched (no subjects)

(3) Most common resolution

  Morgan, T. & Janda, R. (1989): “Frère Jacques”

5

“Happy Birthday to You”: English vs. Korean

Phrases 1, 2, 4: HAP-py BIRTH-day to YOU

Phrase 3: HAP-py BIRTH-day dear SA-RAH

Phrases 1, 2, 4: SENG-il CHU-kah HAP-ni-da

Phrase 3: SA-rang HA-neun JI-EUN-e-eui 6

Experiment:�Syllabic Intensity Tracking through Finger Tapping

7

Objective

Anacrusis/ downbeat perception

perceived intensity contour of syllables

finger tapping while singing

Equipment & Setup

acoustic drum trigger

MOTU audio interface

Computer running Audacity

Praat for analysis

8

Subjects

Complete fluency in English

Complete fluency in Korean

Group I (n=12)

Group II_1 (n=5)

Group II_2 (n=5)

Group II_3 (n=6)

Group II_4 (n=2)

12 Native English speakers with no knowledge of Korean (Group I)

18 Korean-English bilinguals (Group II)

9

Hypothesis

Perceive Anacrusis

Perceive Downbeat

(1)   ACROSS-GROUP singing in English

Native English Speaker Native Korean Speaker

Sing in English Sing in Korean

(2) WITHIN-SUBJECT by a Kor-Eng bilingual

(Offline effects)

(Online effects)

Task

10

Part 1: Short Questionnaire

Part 2: Perform in subject’s primary language (Group I & II)

[1] sing (“la”) [2] sing (“la”) + tap [3] recite (actual lyrics) [4] sing (actual lyrics) [5] sing (actual lyrics) + tap

Part 3: Perform in subject’s secondary language (Group II only)

Identical to Part 2, except with secondary language

Part 4: Final Survey

Summary of Analysis Procedure

11 €

RatioBE =" you"" to"

€

RatioAE =" py""hap"⎛

⎝ ⎜

⎞

⎠ ⎟ "birth"" py"

⎛

⎝ ⎜

⎞

⎠ ⎟ ="birth""hap"

4. Determine RatioA and RatioB

1. Record 5 trials per subject-language

2. Match tap intensity peaks to the syllables in the text

3. Calculate the change in intensity (as ratio) between consecutive syllables:

€

RatioAK ="il""seng"⎛

⎝ ⎜

⎞

⎠ ⎟ "chu""il"

⎛

⎝ ⎜

⎞

⎠ ⎟ =

"chu""seng"

€

RatioBK ="ni""hap"⎛

⎝ ⎜

⎞

⎠ ⎟ "da""ni"⎛

⎝ ⎜

⎞

⎠ ⎟ =

"da""hap"

Result: Across Group comparison over Pattern A

12

Within-Subject Comparison of RatioB ("you"/"to")

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

II_1 II_2 II_3 II_4

Group II Subjects (n= 5+5+6+2= 18)

Med

ian

In

ten

sit

y R

ati

o (

n=

15

per

su

bje

ct)

Sing in English Sing in Korean

13

Result: Within-Subject comparison over Pattern B

14

Perceive Anacrusis

Perceive Downbeat

(1)   ACROSS-GROUP data over Pattern A:

(2) WITHIN-SUBJECT data over Pattern B:

Conclusion

Native English Speaker Native Korean Speaker

Sing in English Sing in Korean

Music & Language: Research Implications

15

1. online effects of lyrics on metric perception of songs

2. offline effects of L1 when singing in L2

3. language fluency may affect the extent to which linguistic stress patterns play a role in the overall beat-strength perception of songs.

Language can play a significant role in the inference of beat accentuation (and by extension, the metric perception) of songs

Singer’s Language Proficiency

Song’s Lyrics Perception of

Metric Accentuation

1.

2.

3.

Thank you

  Acknowledgment -  Jonathan Berger (thesis advisor)

Professor of Music, Stanford University -  Lera Boroditsky (second reader)

Professor of Psychology, Stanford University

  Research Funding -  Major Grant, awarded by the Stanford University

Undergraduate Research Program

16

Selected References   Dell, F. & Halle, J. (2005). Comparing musical textsetting in French and in

English songs to appear in proceedings volume from the first international Paris conference on the typology of metrical forms.

  Mang, E. (2007). Speech-song interface of Chinese speakers. Music Education Research, 9(1), 49-64.

  Morgan, T. & Janda, R. (1989). Musically-conditioned stress shift in Spanish revisited: empirical verification and nonlinear analysis. In Carl Kirschner, and Janet Ann DeCesaris (eds.) Studies in Romance Linguistics, Selected Proceedings from the XVII Linguistic Symposium on Romance Languages. New Jersey: Rutgers.

  Ross, J. (2003). Same words performed spoken and sung: An acoustic comparison. Proceedings of the 5th Triennial ESCOM Conference, Hanover University, Germany.

17