Phrasefinal pitch accommodation effects in Dutch

1
Phrase-final pitch accommodation effects in Dutch

• Judith Hanssen, Jörg Peters Carlos Gussenhoven

2
Phonetic variation
ICPhS XVI, 6-10 August 2007
Background Method Results Discussion
References1/15

• Variation in form of phonological pitch contours
• ? Information structure
• ? Word structure
• ? Syllable structure
• ? Paralinguistics
• ? Time pressure
• Speech rate
• Proximity of accents
• Proximity of phrase boundaries

3
Truncation and Compression
ICPhS XVI, 6-10 August 2007
Background Method Results Discussion
References2/15

• Grabe (1998a, 1998b) Grabe et al. (2000)
• Pitch accent realization in phrase-final position
  in German and English / four varieties of English
• Erikson Alstermark (1972) Bannert Bredvad
  (1975)
• Effects of vowel length on realization of Swedish
  tonal accent 2

Figure 1 Schematic representation of
Compression and Truncation (from Grabe 1998a)
4
Grabe 1998 German and English
ICPhS XVI, 6-10 August 2007
Background Method Results Discussion
References3/15

• Successively less scope for voicing in
  phrase-final position
• Schiefer / Sheafer Schief / Sheaf Schiff / Shift
• Correlate of truncation / compression
• Rate of F0 change

F0 excursion F0 duration
5
Research questions Standard Dutch
ICPhS XVI, 6-10 August 2007
Background Method Results Discussion
References4/15

• What are the effects of time pressure on the
  phonetic realization of contours in phrase-final
  position?
• - effects on rate of F0 change
• 2) Are the realizational effects
  contour-dependent?
• - three nuclear contours
• Fall, Rise, Fall-Rise

6
Experimental set-up Standard Dutch
ICPhS XVI, 6-10 August 2007
Background Method Results Discussion
References5/15
/????/, /???/, /????/, /???/

• Monosyllabic test words
• - long vs. short nucleus
• - sonorant vs. nonsonorant coda
• Three pitch accents Fall, Rise, and Fall-Rise
• Test words embedded in carrier sentences,
  preceded by a context sentence
• 16 speakers of Standard Dutch recorded
• - 5 male, 11 female
• - aged 18-30

7
Example sentence for Fall-Rise
ICPhS XVI, 6-10 August 2007
Background Method Results Discussion
References6/15

• Fall-Rise loom
• A Dit antieke horloge is nog van opa Thijssen
  geweest.
• B Het was toch van opa Loom?
• A This antique watch used to belong to
  grandfather Thijssen.
• B But didnt it belong to grandfather Loom?

8
Acoustic measurements
ICPhS XVI, 6-10 August 2007
Background Method Results Discussion
References7/15

• We chose the same acoustic correlate of
  truncation / compression
• Rate of F0 change
• F0 duration and F0 change
• Fall Rise Fall-Rise

F0 change (in st) F0 duration (in sec)
F0

Time
9
Results (1) Duration of test words
ICPhS XVI, 6-10 August 2007
Background Method Results Discussion
References8/15

• Figure 2 Duration of sonorant rime of test words

n.s.
n.s.
10
Results (2) Mean rate of F0 change
ICPhS XVI, 6-10 August 2007
Background Method Results Discussion
References9/15

• Figure 3 Mean rate of F0 change in Falls (left),
  Rises (mid), and Fall-Rises (right), data grouped
  for phonological vowel length and coda type
• Fall Rise Fall-Rise

11
Results (3) Mean F0 change
ICPhS XVI, 6-10 August 2007
Background Method Results Discussion
References10/15

• Figure 4 Mean F0 change in semitones.

Falls Rises Fall-Rises
Duration in ms.
12
Summary of results
ICPhS XVI, 6-10 August 2007
Background Method Results Discussion
References11/15

• Choice of adjustment strategy in Dutch is
  contour-dependent
• Falls and Rises increase in rate of F0 change
  (compression) and reduction of F0 change
  (truncation)
• Fall-Rises reduction of F0 change with decrease
  in rate of F0 change (pitch range compression)

13
Discussion (1)
ICPhS XVI, 6-10 August 2007
Background Method Results Discussion
References12/15

• Speakers use various adjustment strategies under
  time pressure, that fall in three categories
• - make less effort (1,2,3)
• - work faster (4)
• - take more time (5,6)
• Six possible solutions to time pressure
• 1) Truncation with undershoot of final target
• 2) Truncation with final target present in
  nonsonorant material
• 3) Adjustment of range (range compression)
• 4) Adjustment of speed (rate compression)
• 5) Earlier pitch targets
• 6) Increase duration of sonorant material

14
Discussion (2)
ICPhS XVI, 6-10 August 2007
Background Method Results Discussion
References13/15

• Rate of F0 change as correlate of Truncation /
  Compression
• - Rate of F0 change shows the mean F0 change per
  second
• - It does not provide an average pitch movement
  per word / contour
• Alternative method of acoustic measurements
  possible
• - measure F0 at regular interval (e.g., every 5
  per cent) to obtain clearer picture of pitch
  movements

15
References
ICPhS XVI, 6-10 August 2007
Background Method Results Discussion
References14/15

• Bannert, R., Bredvad, A. 1975. Temporal
  organisation of Swedish tonal accent the effect
  of vowel duration. Working papers 10, Phonetics
  Laboratory, Dep. Of General Linguistics, Lund
  University, Sweden.
• Erikson, Y., Alstermark, M. 1972. Fundamental
  frequency correlates of the grave word accent in
  Swedish the effect of vowel duration. Speech
  Transmission Laboratory, Quarterly Papers and
  Status Report 2-3, KTH, Sweden, 53-60.
• Grabe, E. 1998a. Comparative Intonational
  Phonology English and German. PhD thesis,
  University of Nijmegen. Published in Max Planck
  Institute Series in Psycholinguistics.
• Grabe, E. 1998b. Pitch accent realization in
  English and German. Journal of Phonetics 26,
  129-143.
• Grabe, E., Post, B., Nolan, F., Farrar, K. 2000.
  Pitch accent realization in four varieties of
  British English. Journal of Phonetics 28, 161-185.

16
THANK YOU!
ICPhS XVI, 6-10 August 2007
Background Method Results Discussion
References15/15

• Netherlands Organisation for Scientific Research
  (NWO)
• Joop Kerkhoff
• Frank van den Beuken
• All the participants of the experiment

Acknowledgements