Proceedings Fonetik 2009 - Institutionen för lingvistik

More documents

Recommendations

Info

Proceedings, FONETIK 2009, Dept. of Linguistics, Stockholm UniversityTransient visual feedback on pitch variation for Chinesespeakers of EnglishRebecca Hincks 1 and Jens Edlund 21 Unit for Language and Communication, KTH, Stockholm2 Centre for Speech Technology, KTH, StockholmAbstractThis paper reports on an experimental studycomparing two groups of seven Chinese studentsof English who practiced oral presentationswith computer feedback. Both groups imitatedteacher models and could listen to recordingsof their own production. The testgroup was also shown flashing lights that respondedto the standard deviation of the fundamentalfrequency over the previous two seconds.The speech of the test group increasedsignificantly more in pitch variation than thecontrol group. These positive results suggestthat this novel type of feedback could be used intraining systems for speakers who have a tendencyto speak in a monotone when makingoral presentations.IntroductionFirst-language speech that is directed to a largeaudience is normally characterized by morepitch variation than conversational speech(Johns-Lewis, 1986). In studies of English andSwedish, high levels of variation correlate withperceptions of speaker liveliness (Hincks,2005; Traunmüller & Eriksson, 1995) and charisma(Rosenberg & Hirschberg, 2005;Strangert & Gustafson, 2008).Speech that is delivered without pitch variationaffects a listener’s ability to recall information,and is not favored by listeners. This wasestablished by Hahn (2004) who studied listenerresponse to three versions of the sameshort lecture: delivered with correct placementof primary stress or focus, with incorrect or unnaturalfocus, and with no focus at all (monotone).She demonstrated that monotonous delivery,as well as delivery with misplaced focus,significantly reduced a listener’s ability torecall the content of instructional speech, ascompared to speech delivered with natural focusplacement. Furthermore, listeners preferredincorrect or unnatural focus to speech with nofocus at all.A number of researchers have pointed to thetendency for Asian L1 individuals to speak in amonotone in English. Speakers of tone languageshave particular difficulties using pitchto structure discourse in English. Because intonal languages pitch functions to distinguishlexical rather than discourse meaning, they tendto strip pitch movement for discourse purposesfrom their production of English. Penningtonand Ellis (2000) tested how speakers of Cantonesewere able to remember English sentencesbased on prosodic information, andfound that even though the subjects were competentin English, the prosodic patterns thatdisambiguate sentences such as Is HE drivingthe bus? from Is he DRIVing the bus? were noteasily stored in the subjects’ memories. Theirconclusion was that speakers of tone languagessimply do not make use of prosodic informationin English, possibly because for them pitchpatterns are something that must be learned arbitrarilyas part of a word’s lexical representation.Many non-native speakers have difficultyusing intonation to signal meaning and structurein their discourse. Wennerstrom (1994)studied how non-native speakers used pitch andintensity contrastively to show relationships indiscourse. She found that “neither in … oralreadingor in … free-speech tasks did the L2groups approach the degree of pitch increase onnew or contrastive information produced bynative speakers.” (p. 416). This more monotonespeech was particularly pronounced for thesubjects whose native language was Thai, likeChinese a tone language. Chinese-native teachingassistants use significantly fewer risingtones than native speakers in their instructionaldiscourse (Pickering, 2001) and thereby missopportunities to ensure mutual understandingand establish common ground with their students.In a specific study of Chinese speakersof English, Wennerstrom (1998) found a significantrelationship between the speakers’ abilityto use intonation to distinguish rhetoricalunits in oral presentations and their scores on atest of English proficiency. Pickering (2004)applied Brazil’s (1986) model of intonationalparagraphing to the instructional speech of Chi-102
Proceedings, FONETIK 2009, Dept. of Linguistics, Stockholm Universitynese-native teaching assistants at an Americanuniversity. By comparing intonational patternsin lab instructions given by native and nonnativeTAs, she showed that the non-nativeslacked the ability to create intonational paragraphsand thereby to facilitate the students’understanding of the instructions. The analysisof intonationnal units in Pickering’s work was“hampered at the outset by a compression ofoverall pitch range in the [international teachingassistant] teaching presentations as comparedto the pitch ranges found in the [nativespeaker teaching assistant] data set” (2004,).The Chinese natives were speaking more monotonouslythan their native-speaking colleagues.One pedagogic solution to the tendency forChinese native speakers of English to speakmonotonously as they hold oral presentationswould be simply to give them feedback whenthey have used significant pitch movement inany direction. The feedback would be divorcedfrom any connection to the semantic content ofthe utterance, and would basically be a measureof how non-monotonously they are speaking.While a system of this nature would not be ableto tell a learner whether he or she has madepitch movement that is specifically appropriateor native-like, it should stimulate the use ofmore pitch variation in speakers who underusethe potential of their voices to create focus andcontrast in their instructional discourse. It couldbe seen as a first step toward more native-likeintonation, and furthermore to becoming a betterpublic speaker. In analogy with other learningactivities, we could say that such a systemaims to teach students to swing the club withoutnecessarily hitting the golf ball perfectly thefirst time. Because the system would give feedbackon the production of free speech, it wouldstimulate and provide an environment for theautonomous practice of authentic communicationsuch as the oral presentation.Our study was inspired by three points concludedfrom previous research:1. Public speakers need to use varied pitchmovement to structure discourse and engagewith their listeners.2. Second language speakers, especiallythose of tone languages, are particularly challengedwhen it comes to the dynamics of Englishpitch.3. Learning activities are ideally based onthe student’s own language, generated with anauthentic communicative intent.These findings generated the following primaryresearch question: Will on-line visualfeedback on the presence and quantity of pitchvariation in learner-generated utterances stimulatethe development of a speaking style thatincorporates greater pitch variation?Following previous research on technologyin pronunciation training, comparisons weremade between a test group that received visualfeedback and a control group that was able toaccess auditory feedback only. Two hypotheseswere tested:1. Visual feedback will stimulate a greaterincrease in pitch variation in training utterancesas compared to auditory-only feedback.2. Participants with visual feedback willbe able to generalize what they have learnedabout pitch movement and variation to the productionof a new oral presentation.MethodThe system we used consists of a base systemallowing students to listen to teacher recordings(targets), read transcripts of these recordings,and make their own recordings of their attemptsto mimic the targets. Students may also makerecordings of free readings. The interface keepstrack of the students’ actions, and some of thisinformation, such as the number of times a studenthas attempted a target, is continuously presentedto the student.The pitch meter is fed data from an onlineanalysis of the recorded speech signal. Theanalysis used in these experiments is based onthe /nailon/ online prosodic analysis software(Edlund & Heldner, 2006) and the Snack soundtoolkit. As the student speaks, a fundamentalfrequency estimation is continuously extractedusing an incremental version of getF0/RAPT(Talkin, 1995). The estimation frequency istransformed from Hz to logarithmic semitones.This gives us a kind of perceptual speaker normalization,which affords us easy comparisonbetween pitch variation in different speakers.After the semitone transformation, the nextstep is a continuous and incremental calculationof the standard deviation of the student’s pitchover the last 10 seconds. The result is a measureof the student’s recent pitch variation.For the test students, the base system wasextended with a component providing online,instantaneous and transient feedback visualizingthe degree of pitch variation the student iscurrently producing. The feedback is presentedin a meter that is reminiscent of the amplitude103
Page 1 and 2:
Department of LinguisticsProceeding
Page 3 and 4:
Proceedings, FONETIK 2009, Dept. of
Page 5 and 6:
Page 7 and 8:
Page 9 and 10:
Page 11 and 12:
Page 13 and 14:
Page 15 and 16:
Page 17 and 18:
Page 19 and 20:
Page 21 and 22:
Page 23 and 24:
Page 25 and 26:
Page 27 and 28:
Page 29 and 30:
Page 31 and 32:
Page 33 and 34:
Page 35 and 36:
Page 37 and 38:
Page 39 and 40:
Page 41 and 42:
Proceedings, FOETIK 2009, Dept. of
Page 43 and 44:
Page 45 and 46:
Page 47 and 48:
Page 49 and 50:
Page 51 and 52: Proceedings, FONETIK 2009, Dept. of
Page 101: Proceedings, FONETIK 2009, Dept. of
Page 127 and 128: Proceedings, FOETIK 2009, Dept. of
Page 153 and 154:
Page 155 and 156:
Page 157 and 158:
Page 159 and 160:
Page 161 and 162:
Page 163 and 164:
Page 165 and 166:
Page 167 and 168:
Page 169 and 170:
Page 171 and 172:
Page 173 and 174:
Page 175 and 176:
Page 177 and 178:
Page 179 and 180:
Page 181 and 182:
Page 183 and 184:
Page 185 and 186:
Page 187 and 188:
Page 189 and 190:
Page 191 and 192:
Page 193 and 194:
Page 195 and 196:
Page 197 and 198:
Page 199 and 200:
Page 201 and 202:
Page 203 and 204:
Page 205 and 206:
Page 207 and 208:
Page 209 and 210:
Page 211 and 212:
Page 213 and 214:
Page 215 and 216:
Page 217 and 218:
Page 219 and 220:
Page 221 and 222:
Page 223 and 224:
Page 225 and 226:
Page 227:
Department of LinguisticsPhonetics
show all

Proceedings Fonetik 2009 - Institutionen för lingvistik

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?