Teaching L2 Pronunciation: Tips, Tricks and Tools

  • Published on

  • View

  • Download


PowerPoint PresentationGillian Lordglord@ufl.eduTeaching L2 Pronunciation: Tips, Tricks and ToolsHarvard University August 2016Intersection of topicsFuture directions?We can explore L2 sound systems from different perspectives (Major, 2001):Individual segmentsCombinations of segmentsProsodic/suprasegmental featuresGlobal accentIn terms of exploring instructed acquisition, we have limited ourselves primarily to (1) and (3).Methodological issues (control, sample size, analysis)ISSUES TO FACE:Comprehensibility/intelligibility vs. segmental and suprasegmental dataPerception ~ Production relationshipClassroom exploration > recommendations > empirical outcomes Share work, collaborateCoherent call program unifying good pedagogy with variety of tools, along with sound SLA principles. Discuss with a partner: How important is it to teach pronunciation in the language you teach?Do you (plan to) incorporate explicit instruction in pronunciation in your classes?If so, how? How would we determine success?If not, why not? A brief history(See also: Cenoz & Lecumberri 1999; Elliott 1995; Lord 2008; Pennington & Richards 1986; Terrell 1989; inter alia)Those of us who work in L2 pronunciation tend to always say that were doing it now even though nobody ever does or has, so these quotations are particularly interesting:So why is it that we all think we arent doing it, when for at least the last century folks have been talking about it? Why (not)?World Readiness Standards for Learning Languages (2015)Standard 1.1:Students engage in conversations, provide and obtain information, express feelings and emotions, and exchange opinionsStandard 1.2:Students understand and interpret written and spoken language on a variety of topicsStandard 1.3:Students present information, concepts, and ideas to an audience of listeners or readers on a variety of topics.ACTFL guidelines (2012)Novice Low: have no real functional ability and, because of their pronunciation, may be unintelligible Novice High, Intermediate Low: pronunciation that is strongly influenced by [the] first language, Intermediate Mid: limitations in their vocabulary and/or pronunciation and/or grammar and/or syntax but are generally understood by sympathetic interlocutors accustomed to dealing with non-natives. Distinguished: a non-native accent and/or an occasional isolated language error may still be presentWhich means (Arteaga 2000; Alley 1991; Azevedo 1978; Foote, Holtby & Derwing 2011; Wieczorek 1991)For example:Minimal use of technologies that could helpSo Should we bother?Sociopragmatic consequencesSocial penalty (Oyama 1982, p. 35)Intelligibility and comprehensibility (e.g. Munro & Derwing 1999)Teachers can do much to help students improve their pronunciation Scarcella and Oxford 1994, p. 223Instructors recognize importanceCorts-Moreno 2002; Harlow and Muyskens 1994; Morin 2007 Instructors want more, and more developed, materials Foote et al. 2001Focused instruction outcomes (1):Foundational language classes(Gonzlez-Bueno 1997; Elliott 1995, 1997; Bajuniemi 2015)(Rodrguez-Sabater 2005)(Elliott 1999; Bailey & Brandl 2013)Focused instruction outcomesGonzlez-Bueno (1997): treatment in language classes reduced VOT on all stopsElliott (1995, 1997): explicit instruction on point, place, manner reduced foreign accent on word- and sentence-level readingElliott (1999): instruction improved performance, but not order of acquisitionRodrguez-Sabater (2005): generalized peer tutoring provided greater output opportunities, which improved pronunciation skillsBailey & Brandl (2013): technical and non-technical pronunciation instruction in beginning class, inconclusive resultsBajuniemi (2015): improved pronunciation of /d/ after biweekly instruction in basic language classFocused instruction outcomes (2): Focus on perception(e.g. Flege 1995)Focus on perception (e.g., Flege 1995) Kissling (2012, 2015): explicit instruction aids perception/discrimination as well as productionAusn & Sutton (2010): auditory acceptability judgment test before and after phonetics class reveals increased sensitivity to non-native pronunciations Counselman (2010, 2015): use perception to focus on forms (vowels); comparable to traditional approachesFocused instruction outcomes (3): Phonetics/Pronunciation coursesCastino (1992): traditional instruction regarding point, place and manner of articulation; positive effects of treatmentZampini (1998): significant improvements in VOT during semesterLord (2005): traditional presentation supplemented with acoustic visualization and self-assessmentsLord (2010): delayed effects of phonetics instruction after subsequent immersionDiscuss with a partner: What are the most problematic sounds in the language you teach? Why are they so difficult? What are the biggest L2pronunciation challenges?Not to mention suprasegmental featuresSounds that dont exist in EnglishSounds that are similar to those that exist in EnglishProcesses that occur in English but not in L2Processes that occur in L2 but not in EnglishORTHOGRAPHYSUPRASEGMENTALS!Etc.TIPS AND TECHNIQUESHow?General incorporation suggestionsHammerly (1982): word/phrase imitation Zampini (1994, 1998): L1 interference, orthographyArroyo Hernndez (2009): pedagogical design based on L1-L2 contrasts Mantini (1980): encourage autonomous work, supplement standard lessons with perception and production practiceAlley (1991): include at all levels of instruction; suggests fluency squares/story squaresStringer (1998): Everyday Language PerformanceStokes (2004): strategies for focusing on meaning, communication, and culture while focusing on phonetic formMorgan (2006): phonological flea circus of approaches and ideas to encourage incorporation Step 1: Awareness-raising Step 2: L1 facts Step 3: L2 facts Step 4: Application and practiceFour-step approach (Nibert 2014)Nibert, Holly J. 2014. Disassociating English [, , , ] from Spanish o.Retrieved from D. Eric Holt & Gillian Lord, IGNITE: CASPSLaP Resources.www.ignite-caspslap-2014.weebly.com, {May 6, 2014}.Classroom tips & tricksIn some languages, voicing is contrastive:In some languages it is allomorphic:And in some languages, it is a phonological process:1. Vocal chord functionIngrid KleespiesGorot = city in RussianSay dogs and catsWhile English aspirates /p, t, k/ in a stressed syllable, Spanish (and French, etc.) do not. 2. Voice Onset TimeSay paper and papa(Contrastive) Nasal vowels are found in French, Portuguese, Hindi, etc. 3. Oral vs. nasalsept[st]sainte[st]Sept = 7 in FrenchSainte = saint in frenchOnly difference is in the nasal vowelAlso portugese pao/po, etc. Say sept and saintepao and poAndra FerreiraSounds that are produced with two active articulators at the same time are especially problematic if the L1 produces them as two distinct sounds.4. Co-articulationSherrie NunnCo-articulated consonants or complex consonants are consonants produced with two simultaneous places of articulation. They may be divided into two classes: doubly articulated consonants with two primary places of articulation of the same manner (both stop, or both nasal, etc.), and consonants with secondary articulation, that is, a second articulation not of the same manner.Voglio = italian for I wantBisogno = italian for needCaa = Spanish for cane, or a glass of beerSay canyon ~ caaSherrie: The GL isnt so bad in the middle of a word (for example, voglio) but it is particularly difficult to pronounce the word gli (which functions both a definite article and an indirect object pronoun). I tell them to start to make an L sound, then change their minds and say yee. It works (to some extent) for most students.The difficulty of the GN also depends on the placement in the word. Bisogno is fairly easy, but ingegneria is substantially more difficult. So is gnocchi. I tell them to start making an N sound, and then say yo.5. Vowel heightSay a e i o - uEnglish (and Russian, German, etc.) is stress-timed while Spanish (and French, Italian, Chinese, etc.) is syllable-timed. 6. RhythmSYLLABLETIMEDSTRESSTIMEDEach stress group= 1 beatChristina Overstreet, Theresa AntesUtterance# syllables# beatsA. John drinks beer.33B. John is drinking beer. 53C. Johnny is drinking a beer.73Holly J. NibertRepeat and clapGroups of three simultaneously recite A B C in English all should end at the same timeHolly J. NibertUtterance# syllables# beatsA. ngel bebe sidra.66B. ngela bebe la sidra. 88C. ngela se bebe la sidrita.1010Repeat and clapGroups of three simultaneously recite A B C in Spanish C takes longer than B who takes longer than ALanguages like English distinguish between tense and lax vowels, while in languages like Spanish, all vowels are always tense.7. Vowel tensenessSay seat ~ sitbeat ~ bitdeed ~ didfeet ~ fitPaula GolombekTechnologiesCan make us think outside the curricular box and, by so doing, [] give us an excuse to have more fun than we may have been having with the content we teach (Morgan 2006, p. 119)Automatic voice/speech recognitionVisual acoustic analysisUltrasound imagingInstructional sites/programsSocial media and social toolsOther toolsEnables the recognition and translation of spoken language into text by computers. Automatic speech recognitionhttps://en.wikipedia.org/wiki/Speech_recognition Incorporates knowledge and research in the linguistics, computer science, and electrical engineering fields. Ask Siri:What does the fox sayTell me a scary storyBeatboxVILTSVoice Interactive Training System (VILTS)Rypa (1996), Rypa & Price (1999)Voice Interactive Training System (VILTS)Voice interactive language training system to foster improvement in French comprehension and speaking skillsSpeech recognition allows students to navigate through units using oral communication with various types of system feedback. Pronunciation scoring validated by human ratersSRIUsing spontaneous, unscripted French conversations on various topics, supplemented by excerpts from the French news- paper LeMonde, the VILTS offers the student authentic, unrehearsed French speech as might be heard in an interview. Conversations were collected at various levels on ten separate topics, including domains such as travel, health, education, and politics, from a pool of 100 native speakers in France. The conversations and test form the basis for lesson unit activi- ties. A read version of the conversations was also recorded by the same speakers so that both spontaneous speech and a clearer and (generally) slower version of each conversation is available to the student in the lesson units. As student speech is elicited through a variety of lesson activities, the French speech recognizer listens for oral input that will guide the system response. Each student utterance elicits a response from the system within the context of a particular lesson activity. As the student completes a lesson unit and enough speech has been collected, pronunciation-searing algo- rithms will be employed to compare the student speech with that of native speakers. The system has been designed to elicit a body of student speech sufficient to assure a measure of confidence that the returned pronunciation scores are meaningful. The scores returned to the students will be validated by calibrating the results of the automatic scoring algorithm with the results of human expert ratings of nonnative speech. FLUENcYEskenazi (1999)Carnegie Mellons FLUENcY ProjectLearners speak and receive feedback on pronunciationSpeak, get corrections, listen to self and a native speaker and try again, over and overFeedback on what to correct and how to correct itiSpraakD. Nickolai, http://www.ispraak.com/fm.htmlWeb-based online activity generator Instructor sets up activity by selecting language and including a short text for students to read. Student practices speaking Given immediate feedback, along with audio help from native speakers for problematic wordSiri etc.Limitations of ASR Visual representation of the properties of the speech sound (amplitude, duration, fundamental frequency, etc.)Acoustic analysis/visualization Chun (1998, 2012), Lord (2005), Olsen (2014)Praat www. praat.orgFrequently used with intonation (e.g., Chun 1998, 2012, etc.)Research results encouraging in many languagesLord (2005): phonetics course instruction included visualization and self analysisOlsen (2014): visual feedback paradigm (VFP) in 3rd semester language classes leads to significantly improved productionOther toolsAudacityhttp://audacity.sourceforge.net/ Room EQ Wizardhttp://www.roomeqwizard.com Ravenhttp://www.birds.cornell.edu/brp/raven/ravenoverview.html Talk To MeTell Me MoreHincks (2003), OBrien (2004, 2006)Commerical products incorporating ASR with visualizationAuralogs Talk To Me / Tell Me Moreassesses pronunciation in real time.Hincks (2003), OBrien (2004, 2006)Rosetta Stone bought Auralog in 2013Rosetta StoneGodwin-Jones (2007, 2009), Lord (2015, 2016)Godwin-Jones (2007, 2009)Lord (2015, 2016)Limitations of visualization toolsUses high-frequency sound waves to view inside the body in real timeUltrasound imaginghttp://www.fda.gov/Radiation-EmittingProducts/RadiationEmittingProductsandProcedures/MedicalImaging/ucm115357.htm UltrasoundGick, Bernhardt, Bacsfalvi & Wilson (2008); Ouni (2014); Wilson & Gick (2006)https://www.youtube.com/watch?v=-2mesy5C914 Vocal chord vibrationsTongue and articulator movementsReal timeLimitations to ultrasoundStand-aloneinstructional programsIowa Sounds Projecthttp://soundsofspeech.uiowa.eduFix Your Vowelshttp://uvafon.hum.uva.nl/dirk/ed/ Visualization tool for SpanishTerrell Morgan (OSU) & Christina Garca (SLU)Online tool for asynchronous learning and skill-building in Spanish pronunciationStudents hear the target sound and their own production and see both voiceprintsAutonomous learning allows more agency and independenceTal Como Suenahttp://talcomosuena.spanish.ufl.edu eNunciatehttp://enunciate.arts.ubc.ca Yasuo UoteLimitations to instructional programsSocial networkingBueno Alastuey 2010; Jaramillio Cherrez 2007; Saran, Seferoglu & Cagiltay 2009; Tom 2010;Aslim Yetis 2013; Ducate & Lomicka 2009; Lord 2008; Lord & Harrington 2013; Young 2007Fouz-Gonzlez 2015; Fouz-Gonzlez & Mompean 2012CHATTom (2010): Audiovisual representation in various environments (blog, podcast, SNS); positive outcomes Bueno Alastuey (2010): Use synchronous CMC to help students become aware of their pronunciation Jaramillo Cherrez (2007): synchronous voice chat to improve pronunciationSaran, Seferoglu & Cagiltay (2009): MMS with pronunciation tips, specific words/contextPODCASTYoung (2007): Current practices and future considerations Ducate & Lomicka (2009): pre/post data study with podcast practice; no significant improvement but learners enjoyed experienceAslim Yetis (2013): pre/post with podcasts targeting pronunciation; shows segmental and suprasegmental improvement Lord (2008): phonetics class using podcast, self reflection; treatment correlates to improvement in terms of overall pronunciation ratingsLord & Harrington (2013): acoustic analysis of certain sounds (Spanish ptk, r, rr) as well as attitude; both experimental and control showed improvement on some sounds, not on othersTWITTERFouz-Gonzlez & Mompean (2012): learners receive daily tweets focusing on specific words/problems for English pronunciation; results show improvement, engagementFouz-Gonzlez (2015 dissertation): problematic sounds/words for Spanish speaking learners of English; 3 separate studies using Twitter and other web 2.0; overall positive LimitationsOther tools to have more funVowel worm (1)http://www.cp.jku.at/projects/realtime/vowelworm.html H. Frostel, A. Arzt, and G. Widmer. "The Vowel Worm: Real-time Mapping and Visualisation of Sung Vowels in Music", inProceedings of the 8th Sound and Music Computing Conference (SMC 2011), pp. 214 - 219, Padova, Italy, July 2011.Vowel worm (2)https://byu-odh.github.io/apeworm/ Download the code at https://github.com/byu-odh/apeworm/releases/latest Forvowww.forvo.com LyricsGapswww.lyricsgaps.com Phonetics Bingohttps://goo.gl/mngx0Q Anita SaalfeldMRIVocal Communication Laboratoryhttp://www.carolynmcgettigan.com/#!research/https://www.facebook.com/nadinelavan/videos/10153716278513917/ A cognitive neuroscience lab based in the Department of Psychology at Royal Holloway, University of LondonHello. It's me (and my articulators).https://www.facebook.com/nadinelavan/videos/10153716278513917/ Did you see anything that you might want to try in your classes?Is there something you already do that youd like to share with us?Do you have any questions?Discussion glord@ufl.edu Thank you. http://www.slideshare.net/glord Bibliography