Top Banner
The Phone(cs of Rare Sounds: Produc(on and Percep(on of Aspirated Frica(ves in Sgaw Karen Hugo Salgado [email protected] Jessica Slavic [email protected] Berkeley Linguis6cs Society 40 th Annual Mee6ng February 8, 2014
23

Production and Perception of Aspirated Fricatives in Sgaw Karen

Jan 23, 2023

Download

Documents

Benny Goodman
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Production and Perception of Aspirated Fricatives in Sgaw Karen

The  Phone(cs  of  Rare  Sounds:  Produc(on  and  Percep(on  of  

Aspirated  Frica(ves  in  Sgaw  Karen  Hugo  Salgado  [email protected]  

Jessica  Slavic  [email protected]    

Berkeley  Linguis6cs  Society    40th  Annual  Mee6ng  February  8,  2014  

Page 2: Production and Perception of Aspirated Fricatives in Sgaw Karen

Overview  of  Sgaw  Karen  

2  

•  Member  of  Tibeto-­‐Burman  group  of  the  Sino-­‐Tibetan  languages  (Benedict  1972  )  

•  Spoken  by  1,480,000  people  in  Myanmar  and  Thailand  (Ethnologue,  2012)  

 •  Relevant  phonological  inventory:  /p  t  k/,              /ph  th  kh/,  around  6  tonemes  and  9  vowels  

 •  2  coronal  frica6ve  phonemes  with  contras6ve                  aspira6on:  /sh/  and  /s/  

•     Only  3  languages  in  the  UPSID  database  of  451  

     languages  feature  phonemic  /sh/  (Jacques  2011)    

Myanmar  (Burma)  

Thailand  

Karen  State  

Page 3: Production and Perception of Aspirated Fricatives in Sgaw Karen

Research  Overview  

•  Produc6on  of  /sh/  –  The  aspira6on  feature  on  frica6ves  behaves  differently  than  aspira6on  on  stops.  

–  Aspira6on  and  frica6on  seem  to  compete  for  ar6culatory  6me.  

–  Aspira6on  on  frica6ves  is  subject  to  diminu6on.  

 

•  Percep6on  of  /sh/    –  Speakers  rely  on  different  perceptual  cues  for  aspirated  frica6ves  than  they  do  for  aspirated  stops.  

–  Perceptual  cues  for  aspirated  frica6ves  depend  on  vowel  height.  

3  

Page 4: Production and Perception of Aspirated Fricatives in Sgaw Karen

Part  1:  Produc(on  of  Aspirated  Frica(ves    Research  Findings    •  Aspira6on  (VOT)  on  frica6ves  is  significantly  less  than  stop  

aspira6on.  •  Higher  vowels  increase  stop  VOT  but  decrease  frica6ve  VOT.    •  Higher  vowels  increase  frica6on  dura6on.  •  Nega6ve  correla6on  between  frica6on  and  aspira6on  dura6on  

(Salgado,  Slavic  &  Zhao  2013).  –  Frica6on  dominates  aspira6on  dura6on  in  frica6on-­‐lengthening  

contexts  (i.e.  before  high  vowels).  

       

4  

Page 5: Production and Perception of Aspirated Fricatives in Sgaw Karen

Methodology  •  4  na6ve  speakers  of  Sgaw  Karen  

–  3  female,  1  male;  mean  age:  28    

•  Targeted  6  stops  [p  t  k  ph  th  kh]  and  2  frica6ves  [sh  s]    •  Segments  followed  by  5  vowel  contexts:    

–  high  [i  u],  mid  [e  o],  and  low  [a]  –  represented  by  2  tones  (tone  does  not  seem  to  effect  aspira6on)  

 •  220  randomized  s6muli  presented  in  the  Karen  script  

5  

Page 6: Production and Perception of Aspirated Fricatives in Sgaw Karen

Acous(c  Analysis  in  Praat  

[s]   [sh]  

Coronal  frica6on:  turbulent  airflow  of  coronal  origin  Aspira6on:  turbulent  airflow  of  glojal  origin  (Johnson  1997)  

frica6on   frica6on   aspira6on  

6  

Page 7: Production and Perception of Aspirated Fricatives in Sgaw Karen

Aspirated  frica(ve  VOT  very  short  

7  

Stop  aspira6on  (VOT)  significantly  greater  than  frica6ve  aspira6on.  Frica6ve  aspira6on  (VOT)  near  percep6on  boundary  for  stops,  Johnson  (1997).  

Page 8: Production and Perception of Aspirated Fricatives in Sgaw Karen

High  vowels  increase  stop  VOT,  decrease  frica(ve  VOT  

8  

Page 9: Production and Perception of Aspirated Fricatives in Sgaw Karen

High  vowels  increase  frica(on  dura(on;  frica(on  the  force  behind  short  VOT  

No  significant  difference  between  total  dura6on  of  /sh/  and  /s/.  Aspira6on  unstable  –  subject  to  diminu6on.  

9  

Page 10: Production and Perception of Aspirated Fricatives in Sgaw Karen

Perceptual  Implica(ons  

•  VOT  in  /sh/is  significantly  shorter  than  in  aspirated  stops.  

•  VOT  in  /sh/  is  actually  closer  to  that  of  unaspirated  stops.  

•  Is  VOT  the  main  cue  to  iden6fy  aspirated/unaspirated  frica6ves  in  Sgaw  Karen?  

10  

Page 11: Production and Perception of Aspirated Fricatives in Sgaw Karen

Part  2:  Percep(on  of  Aspirated  Frica(ves    

Research  Findings  

•  Cues  to  iden6fy  /sh/  are  different  from  cues  to  iden6fy  aspirated  stops.  

•  Speakers  rely  on  mixed  cues  to  dis6nguish  /sh/  and  /s/  depending  on  vowel  context.    

•  VOT  used  to  iden6fy  /sh/  in  low  vowel  contexts.  •  Vowel  quality  used  to  iden6fy  /sh/  in  high  vowel  contexts.  –  c.f.  aspirated  affricates  (Clements  &  Kha6wada,  2007)  

 

11  

Page 12: Production and Perception of Aspirated Fricatives in Sgaw Karen

2  tests  to  determine  perceptual  cues      

•  VOT  con(nuum  test  –  Aspirated  stops  and  frica6ves  –  3  VOT  Levels:    

•  short  (near  0  ms)  •  medium  (50  ms)/(28ms  for  /sh/)  •  long  (100ms)/(54ms  for  /sh/)  

–  2  Vowel  Heights:  /a/  and  /i/  

 

•  Conflic(ng  cues  test  –  Cross-­‐splice  stops/frica6ves    

 and  vowels  of  different    categories  

–  2  Vowel  Heights:  /a/  and  /i/  

12  12  

Page 13: Production and Perception of Aspirated Fricatives in Sgaw Karen

Percep(on  Experiment  

 [sho]          [so]                      

   

 [pha]          [pa]        

 [shi]            [si]      

 [sha]              [sa]  

13  

Page 14: Production and Perception of Aspirated Fricatives in Sgaw Karen

Long  and  medium  VOT  cue  aspira(on  on  stops   /t/  

/k/  

/p/  

14  

Page 15: Production and Perception of Aspirated Fricatives in Sgaw Karen

[i]  requires  longer  VOT  to  cue  aspira(on  

/s/  

15  

Page 16: Production and Perception of Aspirated Fricatives in Sgaw Karen

Vowel  Affilia(on  does  not  affect  aspirated  stop  percep(on  

/p/  

/t/  

/k/  

16  

Page 17: Production and Perception of Aspirated Fricatives in Sgaw Karen

Quality  of  vowel  cue  for  high  vowel  contexts  

/s/  

17  

Page 18: Production and Perception of Aspirated Fricatives in Sgaw Karen

Rarity  of  aspirated  frica(ves  explained?  

•  Difficult  to  produce:  VOT  in  aspirated  frica6ves  is  significantly  shorter  than  VOT  in  aspirated  stops.    

•  Aspira6on  and  frica6on  compete  for  ar6culatory  6me.  

•  VOT  in  frica6ves  is  further  shortened  before  high  vowels.    

Page 19: Production and Perception of Aspirated Fricatives in Sgaw Karen

Rarity  of  aspirated  frica(ves    explained?  

•  Difficult  to  perceive:    VOT  is  the  main  cue  to  iden6fy  aspirated/unaspirated  stops.    

•  Percep6on  of  aspirated  frica6ves  depends  on  mixed  cues:  VOT  in  low  vowel  contexts,  vowel  quality  in  high  vowel  contexts.  

Page 20: Production and Perception of Aspirated Fricatives in Sgaw Karen

Future  Research  

•  What  quali6es  in  the  following  vowel  are  relevant  for  aspira6on  percep6on?  – Breathy  voice?  (H1,  H2)  – Vowel  length  

•  Are  there  cross-­‐linguis6c  allophonic  rules  that  treat  aspira6on  differently  in  varying  vowel  contexts?  

20  

Page 21: Production and Perception of Aspirated Fricatives in Sgaw Karen

Thank  you!  

•  We  are  deeply  grateful  for  the  generous  assistance  of  our  Na6ve  Speaker  Consultants.  

•  Special  thanks  to  David  Mora  Marín,  Jennifer  Smith  and  Ellioj  Moreton.  

•  Special  thanks  to  our  UNC  colleagues  in  P-­‐Side  and  K-­‐Side.  

21  

Page 22: Production and Perception of Aspirated Fricatives in Sgaw Karen

References    Blevins,  J.  (2004).  Evolu2onary  phonology:  The  emergence  of  sound  pa;erns.  Cambridge:  Cambridge                      

 University  Press.  

Clements,  G.N.,  &  Kha6wada,  Rajesh.  (2007).  Phone2c  realiza2on  of  contras2vely  aspirated  affricates  in  Nepali.  Proceedings  of  the  16th  Interna6onal                

 Congress  of  Phone6c  Sciences  (ICPhS  16).      Chaida,  A.,  &  Nirgianaki,  E.,  &  Fourakis,  Marios  (2009).  Temporal  characteris6cs  of  Greek  frica6ves.    

       Paper  for  the  9thInterna6onal  Conference  on  Greek  Linguis6cs,  Chicago,  Illinois.  

 Ethnologue  hjp://www.ethnologue.com/show_language.asp?code=ksw      Jacques,  G.  (July  01,  2011).  A  panchronic  study  of  aspirated  frica6ves,  with  new  evidence  from  Pumi.    

         Lingua,  121,  9,  1518-­‐1538.  

 Jongman,  A.,  Wayland,  R.,  &  Wong,  S.  (January  01,  2000).  Acous6c  characteris6cs  of  English              

       frica6ves.  The  Journal  of  the  Acous2cal  Society  of  America,  108,  3,  1252-­‐63.    Johnson,  K.  (1997).  Acous2c  and  auditory  phone2cs.  Cambridge,  Mass:  Blackwell  Publishers.      Kim,  H.  (January  01,  2001).  A  phone6cally  based  account  of  phonological  stop  assibila6on                        

     Phonology,  18,  1,  81-­‐108.    Kuan-­‐Yi,  C.,  &  Li-­‐Mei,  C.  (June  01,  2008).  A  Cross-­‐Linguis6c  Study  of  Voice  Onset  Time  in  Stop                              

     Consonant  Produc6ons.    215-­‐231.   Salgado,  Slavic  &  Zhao.  (2013).  The  produc6on  of  aspirated  frica6ves  in  Sgaw  Karen.  Studies  in  the  Linguis2c  Sciences:  Illinois  Working  Papers.  148-­‐161.    Yavas,  M.  (2009).  Factors  influencing  the  VOT  of  English  long  lag  stops  in  Interlanguage  phonology,  in  M.  Watkins,  A.  Reuber  &  B.  Bap6sta  (Eds.)  Recent      

 Research  in  Second  Language  Phone2cs/Phonology:  percep2on  and  produc2on.  Cambridge  Scholars  Publishing.  244-­‐255.      

22  

Page 23: Production and Perception of Aspirated Fricatives in Sgaw Karen

Conclusions  

•  Difficult  to  produce:  VOT  in  aspirated  frica6ves  is  significantly  shorter  than  VOT  in  aspirated  stops.    

•  Aspira6on  and  frica6on  compete  for  ar6culatory  6me.  •  VOT  in  frica6ves  is  even  shorter  before  high  vowels.    •  Difficult  to  perceive:    VOT  is  the  main  cue  to  iden6fy  

aspirated/unaspirated  stops.    •  Percep6on  of  aspirated  frica6ves  depends  on  mixed  cues:  

VOT  in  low  vowel  contexts.  Vowel  Quality  in  high  vowel  contexts.  

23