Report on the Phonetics Teaching and Learning Conference (#PTLC2015) at UCL. August 5-7th, 2015 - Part 2

My previous post briefly commented on the sessions I attended at PTLC 2015 on Day 1. I am now going to report on a very busy day at the conference, Day 2, before it all flies away from my short-term memory and all my teaching duties take over! 

(My usual disclaimer: all misinterpretations of the presentations are my own, and totally unintentional)

Day 2: August 6th

The opening plenary was by Dr. Eleanor Lawson, who struck me as sooo knowledgeable and so young! She made a fascinating report of a few studies on Scottish postvocalic /r/, and what different types of measurements (impressionistic, MRI, UTI) contributed to the study of coarticulation. The presentation continued with the description of the whole development process for these two amazing online resources: Seeing Speech and Dynamic Dialects. It was, simply, an amazing presentation.

And after that, we had the fabulous experience of trying ultrasound imaging on our own tongues! 

Trying my Spanish alveolar trill! :D Sooo exciting!

The second session of the day started with a very entertaining presentation of teaching ideas for laboratory phonetics by Timothy Mills, Karen Pollock and Benjamin V. Tucker from the University of Alberta. One of their techniques involved getting students to create their own "Frankeintracts" of the vocal tract, making use of any material of their choice (apparently some students even attempted models that could actually utter sounds!). Other very interesting activities consisted in the students' plotting of their own vowel formants and a few designs of tests for the perception of neutralisation of items such as "latter/ladder". 

Mercedes Cabrera-Abreu and Francisco Vizcaíno-Ortega were up next (represented by Mercedes on this ocassion), and their presentation was a discussion of their classroom assessment tasks for their courses on acoustics, and the results obtained. From spectogram and waveform recognition tasks to the actual hand-drawing of spectograms, students were assessed on numerous aspects of acoustic phonetics, with different levels of success. I found the tasks particularly interesting (and difficult, given my really basic knowledge of acoustics!), and this presentation opened my mind to a other ways of testing.

I, Marina Cantarutti, was the following presenter. I discussed a very humble classroom experience on intonation teaching through speech genres (the lecture genre, in this case) and on my students' treatment of tonality, tonicity and tone in connection to thematic structure in a pre- and post-instruction tests. I have basically found that my students actively assigned different degrees of relevance to the various thematic elements in the text given by either conflating Themes and Rhemes in the same IP (lowering the relevance and possible contrastive value of the thematic element) or by treating them obliquely. I also made a point that contrary to my own expectations, students initially produced more cases of transfer from Spanish in terms of their treatment of focus, rather than of tone.

After lunch, two attempts at a group picture, and a tour round the labs and the library, we were all ready for more.

The whole #PTLC2015 bunch! (Credits: PTLC, FB page)

Smiljana Komar from the University of Ljubliana introduced her results in perception tasks for the fall-rise tone. She found some interesting cases of mistaken perception of fall-rises for falls, and then for rises. Her findings in a way appear pretty similar to the ones I believe we would find over here in Buenos Aires, if we tried the same tests, which makes the whole thing really intriguing, given that we have different L1s!

Yusuke Shibata, Masaki Taniguchi and Tamikazu Date presented an experience with junior high teacher and students, connected to notions of tonicity and focus. They have found these features to be highly teachable, and they expect to be able to "persuade" and also train teachers towards the active integration of pronunciation and intonation work at schools.

Junko Sugimoto and Yoko Uchida carried out an analysis of the government-approved ELT books in Japan in search of pronunciation tasks and training. They have found that there were activities connected to vowels, consonants and well as articulatory explanations and resources on phonics and Katakana. In my opinion, the books they analysed presented a number of very interesting contents and tasks, and they far exceeded the number of activities and pronunciation training available in the EFL textbooks that we have on this side of the world.

Nikola Paillereau presented a comparison between some specific Czech and French vowels as produced by students acquiring French as L2. The focus on this presentation was the assessment of L2 vowels using a piece of software called VisuVo -which, unfortunately, is not open to the public yet-, which was designed by the presenter and a collaborator . The program allowed for measurement and plotting of vowel formats and comparison of other variables across speakers and in an intra-speaker manner. 

Rungpat Roengpitya discussed the design of different Phonetics courses at different departments at her university in Thailand. I found it particularly interesting that the inclusion of Phonetics for training in some medical sciences, such as Dentistry, was aimed at helping the future professionals become aware of how they can improve a patient's quality of life by knowing how their intervention may affect speech.

Pekka Lintunen and Aleksi Mäkilähde (represented by the former on this occasion) carried out a very interesting study regarding what students prefer, like and find motivating about the Phonetics courses they attended. Their survey revealed that students find accents and intonation topics more engaging than other themes in the course. Part of their study also assessed whether students' view of Phonetics as highly benefitial to their future career had changed, and in most cases, students agreed that Phonetics was necessary for their professional development. There were a few caveats and self-objections that the presenter made to the survey and its delivery, but it was overall a very interesting, and easily replicable study, worth further thought!

The closing prenary was by Professor David Deterding, and it was aimed at discussing misunderstandings and the role of pronunciation for intelligibility. We participants had a lot of fun decoding many instances of English as an L2/FL speech from different locations (Brunei, Nigeria, among a few others), and in the end, it turned pretty challenging to make sense of many words. (This may seem quite obvious and ordinary for people teaching in multilingual environments, but in my teaching setting, where most of my students' L1 is Spanish and where English is only used in the classroom, activities like these really open up your mind!) . Deterding's talk included a review of the Lingua Franca Core, and some comments regarding attempts at revised versions in different environments. Apart from the well-known objection to stress-timed rhythm as a feature making speech less intelligible from an international perspective, there were a few comments regarding the role of consonant clusters in the blurring of comprehension at word level. Event though David did not perhaps tackle upon this explicitly in his talk, later personal communication made it clear that of course, we need to make a distinction between aims we may have regarding perception, and those for production. We all agree that perception and exposure to all accents of English, including "international Englishes" is essential if we want our language teaching training to be enabling and empowering for communication.

***

This is the end of Day 2 (because, of course, I will not report on the wine-tasting session!). Just in case: I'm just the messenger here, so I am merely reporting on the sessions, and adding a few comments, but of course, any objections or remarks on the presentations should be addressed to the authors themselves (do note that I have, in most cases, added links to the speakers' institutional affiliations!).

 I'll be wrapping up my discussion of PTLC on my next post on Day 3, which will probably be out during the weekend. Thanks for bearing with me!

Report on the Phonetics Teaching and Learning Conference (#PTLC2015) at UCL. August 5-7th, 2015 - Part 1

I have just attended another fabulous conference, this time, on the teaching and learning of Phonetics at University College London. I was also blessed with the opportunity of presenting my own paper! PTLC takes place every two years, and as with other conferences the papers submitted are subjected to blind review by the academic commitee (hopefully the proceedings will be published soon!). In the meantime, you can read about the conference and check previous proceedings here.

In this post, I will be talking about some of the presentations I would like to comment on (as usual, any misinterpretations are my own). If you want to get a deferred "feel" of what the conference was like, you may want to see the Storify collection of live- tweets that Jane Setter put together (scroll to the very bottom of this post - and yes, there was some hectic tweeeting!)

This first post is on Day 1. 

Day 1: August 5th

The conference kicked off with University of Reading's Jane Setter's presentation of her "flipped learning" experience in her Phonetics courses at graduate and undergraduate levels. This methodology basically consists in making the most of class-time for practical and problem-solving tasks while assigning videotaped lectures and practical tasks with key as homework. There were some interesting references to the well-known Bloom's taxonomy, and also to Dale's 1969Cone of Experience, a thought-provoking approach to types of learning tasks that may contribute to the increase of retention types. Jane discussed the results of her assessment tasks with the flipped and "unflipped" (?) classes, with a marked improvement for students in the former, except for the dictation tasks. (Dictations and decoding tasks are always a problem, if you ask me! This does deserve a whole new post!)

Shawn Nissen from Brigham Young University also reported on a blended learning experience with large Phonetics classes (for Speech Therapy trainees, if I remember correctly) consisting of, among other things, lots of self-paced theoretical and practical tasks with key, leading to online assessment tasks (of which an 80% success grade should be achieved). I found it particularly interesting, and challenging as well, that many of the materials for decoding included audio files containing noise levels, and also disordered speech. There were also some true and realistic remarks worth recalling, such as the fact that Phonetics teaching needs to be orderly and it needs a "unique learning environment"and set of techniques we may not find when teaching other subjects.

Alice Yin Wa Chan introduced some of her own ideas to bring variety into the Phonetics classroom, including the use of water bottles as an analogy to draw students' awareness of the cavities, shapes, amplification in the vocal tract. She also found ways of connecting students' cultural practices (like the use of acronyms to describe their relationship status) to minimal pairs, and a reference to Peter Roach's comparison of allophones to different styles in handwriting.

Gwen Brekelmans discussed the results of her study on a group of Dutch learners of English at University and compared the effects of lack of explicit pronunciation instruction in two groups, one group of learners who had taken their "year abroad" experience, and others who had instruction for two years but were not trained during their third year. Once instruction had stopped for the second group of learners, subjects showed a clear deterioration of their foreign accent, especially in vowels. Those students spending a year abroad did show a bit more improvement. In all cases it was found that pronunciation of English as an L2 did not remain fully stable after instruction (which is exactly what happens with many of my teacher trainees once they graduate, sadly. Lots of pondering to do in this respect. It is definitely something worth giving a thought...)

Eva Estebas made a truly interesting presentation (fitting timely and perfectly with my current personal interests!) on an adaptation of ToBI and the School of London's nuclear configurations to teach intonation and help learners work autonomously at a graduate distance programme (big challenge, if you ask me!). Eva called this special concoction "TL_ToBI" (Tone and Break Indices for Teaching and Learning), and I would say she collected the best of both approaches, focusing on the connections between metrical and tonal structure, and the notation and visual aids for ToBI, and the tune description and nuclear configurations of the British School. Estebas reported an improvement in students' perception and production of English Intonation after employing this new approach.

***

Here ends my Day 1 report on PTLC. More on this great conference coming up soon!

In the meantime, take a peek at this wonderful collection of tweets!

Report on the panel "Prosodic Constructions of Dialog", at the 14th International Pragmatics Conference - Part 2

In my previous post, I reported the first two sessions of the panel "Prosodic Constructions of Dialog" at the International Pragmatics Conference. Here's the second and final part of this "racconto".

Session 3: Prosodic Constructions and Actions

Beatrice Szczepek-Reed discussed the case of the German version of "yes, but"- " ja aber" as a tactic or strategy, and its different phonetic manifestations in interaction, which appear to project different affiliative stances. As second pair parts, they may be realised in a continuum going from "ja aber", with the first element comprising a first TCU and the second element carrying a pitch accent and  lengthening; to " jaber", compressed, with the presence of glottalisation (also for some of the tokens in the first group, btw).  "Ja aber" was found to express disaffiliation through disagreement or qualification, whereas "jaber" "re-does" disffiliation, as either previous rejection or a lack of uptake. There was a very interesting remark about the role of the "pre-front field " (Auer, 1996), the first position at the beginning of a TCU, which is the locus for new "creations" in language and the conversion of many items into discourse markers (this leads me back to the presentation on Pragmatic Markers by Romero-Trillo on the " Pragma-Discourse..." on Monday, and the role of continuers as interpersonal and textual markers, as seen by SFL). I found Beatrice's remark that "in order to study a big thing you have to work on a small thing" particularly inspiring!

The following presentation by Rasmus Persson surveyed different roles and prosodic constitutions of repetition in French, in particular, of other-repeats. The presenter reviewed some of the characteristics of French, especially in terms of stress, before discussing the two prosodic formats of the cases of other-repeats found in his data: early peak with a salient secondary accent, and a late rise fall, with a salient primary accent. Whereas the first configuration has been associated with the acknowledgment of receipt, "registering" previous talk, the second case was found to initiate repair, especially implicating some sort of "problem" with the content of the talk being repeated. The presenter wraps up his discussion by making a very important point for the study of phonetics in interaction: "the interactional function of prosody can be understood by taking into account the action the prosodic design contributes to, what a turn as a whole is supposed to do".

Maria Ibh Crone Aarestrup and Kerstin Fischer discussed an experiment on the way native and non-native speakers perceive greetings as produced by robots. The presenters reviewed some bibliography on the intonation of greetings, such as Wells (2006), Tench (1996) and then described the experience, which included synthetic productions of greetings produced by three different robots , which were then rated by native and non native speakers of English.  From what l have understood, native speakers found the high wide falling tone on greetings more inviting, which was, to me, at least, expected, though the presenters, were apparently expecting those with a rise to do the trick. But then, it may be a misunderstanding on my part.

The discussion session was really fruitful, with comments on the notion of prosodic " constructions" as actual processes unfolding in time, fluid and flexible, vs the perhaps "rigid" association that could be wrongly made with the term. There was also another true remark about the fact that when it comes to the analysis of the enactment of actions through prosodic constructions, there is always ambiguity and gradience to be wary of.

Session 4: Prosodic Constructions and Gesture

David House reported on a number of experiments carried out with Margaret Zellers to see connections between gesture and prosody. The audience had fun watching the videos of the interactants looking like "ants" with their antennae and a few other props recording the speakers' every move. The presenter reviewed a few of the already-identified relationships between prosody and co-speech features, such as prominence marking, phrase connection and demarcation, dialogue regulation and turn-taking, and reported that in English, it is the intonational cue that prevails. This study was concerned with turn boundaries, and some of the findings were that turn-holding practices show lengthening of both prosodic features and gesture, and in the case of turn-yielding boundaries, the end of gesture precedes the end of talk spurt before transition. This was corroborated manually and through automatic methods, but not yet tested for statistical significance.

Romero-Lopes, Del Re and Dodane (represented by the latter), discuss their findings regarding prosody, gesture and the acquisition in infancy of the Brazilian "preterito perfeito simple" tense. They found that sound lengthening and gestures were synchronised when the conjugated verb was produced. (I may not have been faithful to their results, I am afraid, so my apologies for any misinterpretation).

The final presentation was by Richard Ogden and Verónica González-Temer, on the particle "mm" in Chilean Spanish, and it was aimed at exploring its function and different vocal and non-vocal modalities. The data was very interesting, as it consisted of elicited situations where speakers were asked to taste food they had never had before, and interact with another person to reach an agreement on the rating and possible ingredients of what they have just tried. The tokens were coded according to the taxonomy by Gardner (2001) and classified in terms of action types. The presenters have found that as response tokens, "mm" was uttered with low falls, glottal stops, nods and quite short in length. Non-response tokens showed greater degrees of variation (especially in terms of gesture) depending on whether they were enacting incipient speakership, gustatory or lapse terminator actions.

The final discussion was as enriching as for the previous sessions, including the acknowledgement that there are different semiotic systems interacting in a parallel fashion, that gesture should be part of the analysis, and that it makes sense to see how participants themselves make sense of the prosodic constructions themselves. This, of course, marks a break from other linguistic traditions, as it drives the focus away from what is normative, into what is descriptive.

I hope I have delivered a faithful account of the sessions of this fantastic panel, and my apologies for any misinterpretation and typos (wrote this post at the airport!). It's been a true blessing for me to have attended it.

Report on the panel "Prosodic Constructions of Dialog", at the 14th International Pragmatics Conference - Part 1

Report on the panel "Prosodic Constructions of Dialog", at the 14th International Pragmatics Conference - Part 1

I am currently " conference-crawling" and in an effort to be faithful to my second linguistic love, Pragmatics and Discourse Analysis (Pragma-Discourse), I spent six days at the International Pragmatics Conference .  I will be reporting on a fantastic panel I attended on Tuesday, in two installments .

The usual disclaimers: I am writing from my tablet so I am afraid there won't be many "special effects" and there will probably be a million typos. Plus, any misunderstanding of theoretical views is my own (attributable to jet lag?).

Last Tuesday I had the fabulous experience of attending twelve talks as part of a panel that dealt with " prosodic constructions", that is:

"Units of (phonetic and prosodic) organization that unite elements of linguistic form with elements of meaning, including affiliation and sequence management" (Ogden, 2010 as quoted by Szczepek-Reed, slide 2)

The panel was organised by Nigel Ward, Richard Ogden, Oliver Niebuhr and Nancy Hedberg, and you can read about it on this website .

I would like to briefly report what I have heard, and include some personal remarks about the presentations and the whole experience, starting with the fact that the organisers came up with a fantastic idea of not only allotting the usual 5 min Q&A, but also a whole 15-minute slot for open discussion with the audience. This really made all the difference, in comparison to other panels l have attended .

First session: Compositional and Constructional Aspects of Prosody
Yi Xu et al opened the panel with a very passionate presentation of a piece of software (PENTAtrainer2 - this link goes to the first version) that allows the decomposition of different prosodic elements and their reassembly, by "extracting global patterns as sequence of local targets". This is done through easy annotation and coding of issues such as stress, pitch height and slope, "focus" (his use of the word in fact is what we understand as nucleus placement that is, phonetic rather than informational, if I understood his account correctly - he does not appear to make a distinction between  the informational and intonational choices at this stage), modality (something like voice quality), among other things. The tool and system presented comes as a result of extensive research and literature review.

The second presentation, by Nigel Ward, addressed "narrow pitch regions" (the Bookended-Narrow Pattern), which the author associated with a " squished down pitch range" that appears to take place in some occasions in which speakers express contrast, complaint, grudging admiration and contradiction (Sorry, but I got a sudden flashback to O'Connor and Arnold, especially because of the wyrd "grudgingly"). This pitch range area is problematic for analysts as it is too subtle, and there is an interplay of multiple prosodic constructions which are superimposed. This pattern is presented as being used to " introduce specific personal knowledge and invite the interlocutor to consider it". (Aug 28th, update: The presenter has shared with me teaching resources on his research on prosodic constructions: http://www.cs.utep.edu/nigel/patterns/)

This presentation opened up a very interesting debate regarding the definition of functions, prescriptiveness, the role of context in determining that these prosodic choices may "stand for" and the establishment of categories (Whose? The analysts'? The participants'?) Clearly this is an issue that marks interactivists apart from other approaches (and the one I enjoy the most; the actual possibility of describing what actions speakers appear to be implementing through prosody, and how,  instead of presenting generalisations - which, on the other hand, I have to do because of my job!)

The last presenter in this session was Oliver Niebuhr, who focused on Laboratory data, and the factors that affect elicitation (including issues like time of day and type of font used in the text, apart from the usual factors we may know about). The presenter proposed new elicitation techniques (under the name INSPECT) to enable the retrieval of data that could resemble the characteristics of spontaneous speech more closely. Those of us "interactionalists" (in the making,  at least, for myself) will always prefer naturally occurring data, but it was granted by the author that laboratory speech is  a "natural reaction to a controlled environment".

The ensuing discussion was truly enlightening, a bit fiery for some (and yet everyone was so respectful!). One of the contributions by the discussant, Richard Ogden, was very clear in terms of the fact that this this form- function issue could be delimited in terms of the difference between something having a phonetic exponent, versus phonetic structure having meaning.


Session 2: Prosodic Constructions and Turn-Taking

Jan Gorisch et al describe an experience of replication of a study by Kurtic et al 2013 on overlapping talk and whether the instances identified in the corpus of multiparty conversation could be said to be competitive or non-competitive. In this occasion,  the tokens were rated in terms of (non) competition by different raters, and also subjected to a Decision Tree.  It had been established in previous studies that prosodic matching was related to interactional alignment, and this presentation finds a relation between position in the previous TCU where the overlap starts, the duration of overlapping talk, and the pitch span with issues of turn (non) competitiveness.. (Found this presentation particularly interesting, since it is closely related to my MA thesis, BTW)

Stevanovic and Lennes present a study that is aimed at connecting pitch matching -absolute and relative- to the notion of agency. This latter concept, apparently related to issues of sequential organization (later connected to levels of passivity-activity of the interlocutors, and also discourse control, in a more informal description of the concept) was presented as connected to matching and non-matching of speech. The discussion that followed introduced a few other issues that could affect levels of agency, such as backchannel practices that may require a lower level of agency.

De Ruiter commented on a 2006 study which over the years has become quite polemic (but which could not be refuted, according to the presenter), regarding the apparent lack of speaker's reliance on pitch to decide on possible TCU ends. The experiment consisted in presenting subjects with natural and synthetic turns, first with words and pitch, then with "blurred" words, then noise, for them to press a button when a TCU appeared to be reaching a potential end. The ensuing discussion introduced the possibility of other variables that may "do the trick" at TCU boundaries, such as lentghening and rhythm.

Not only was it a fabulous panel, but it was also a sort of a dream come true for me, as you can imagine!

Coming up soon: part 2, and of course, PTLC!

Talk like Hawking: Teaching with a synthetic voice

Hiya! I have been pretty busy and also a bit disappointed in life lately (soo dramatic!), but blogging has always been good medicine, so I'm back! 

***

I have always found voice matters interesting, and I do pay a lot of attention to people's voices in general. They have the power of generating feelings, immediate gut feelings, in me. 

For many years, I was concerned about my own voice, as it would invariably become hoarse and then turn to a whisper at least twice or three times a year. I was teaching over 55 periods a week and as we all know, our teaching activity is a 100% dependent on our voice (there's a limit to emergency video sessions one can plan when your voice's "on the blink"!). So I was basically "killing my voice softly".

Things changed for the better when I started training in vocal technique, and I will always be grateful to Carlos and Liliana at Instituto de la Voz for having saved me from potential hiatus and nodule problems. And I have now got the huge responsibility of passing on these techniques and voice care tips to my teacher trainees, which I try to do on a regular basis. 

Now. This whole babble is related to today's blog post, since in spite of all my care, my voice was gone yesterday due to an irritating bout of flu or something (and the winter temperatures hitting our country like we haven't seen for a number of years!). With several classes a day to teach, all pronunciation-related, how do you keep going?

Well, the obvious answer is: "You don't. You go home, you see a doctor, you get some rest for a few days till you recover". And that would be the right answer. But it's not always possible. So here's what I did yesterday.

***

Teaching Phonetics the Hawking-style

We are all familiar with Stephen Hawking. We all know what a miracle it is that he can share all his wisdom and knowledge with us through a computer. Therefore, I used him as an inspiration to try to teach my class without having to strain my voice.

So here's what I have done:

1) I typed part of my lecture at home, on my tablet (It runs an Android Kit Kat operating system).

2) I downloaded the following app: Talk, and selected English (GB) as the output voice. (I was in a bit of a hurry, so this is the first app that worked for me, but I will be comparing it with others below)

3) I tried the app out, and noticed that to make it less "robotic", I had to "teach" the system to read the way I wanted it to read. That is, I had to input extra punctuation marks, such as "" and ; to allow the machine to chunk units apart or produce some basic intonation patterns which would allow my audience to follow more clearly.  I also had to simplify my clauses and exploit thematic marking on the text to make sure my synthetic voice would signpost topics and paratones more clearly to my students.

4) I gave my tablet Internet access via my phone, using it as a router, since this app needs a working connection, and of course, I connected my tablet to my portable speakers.

The immediate result? Giggles, of course! It was fun. I got called "Marina Hawking". But it was, of course, not the same as being able to respond to students' queries in real time. I eventually had to replace the use of my tablet for the whiteboard when I needed to reply to students' queries on the spot. 

I wanted to brag and I finished my lesson by typing in "That's all, folks!", but the awful rendering by the system (giving "folks" a separate IP and using a fall) spoiled all the fun! (Try it!)

However, I was teaching a theoretical class, so the app did the trick. I had a Lab lesson right afterwards, and it was really difficult to succeed in this way. I was not in a position to deliver feedback on the spot, since obviously, my app would not read IPA symbols, and the need for synchronous comments while the student is reading out, or speaking, does not match my typing speed. In this particular case, gestures and the whiteboard did the job. Sort of. My teacher assistant allowed the class to happen, to be honest!

(Had I had a data projector, I would have projected the text my students were reading, and I would have highlighted their slips and mistakes on the spot, to allow them to at least see what I was trying to comment on! I could have also projected some vids and audio files from some of the pronunciation apps I reviewed in the past.)

So all in all, a text-to-speech act may save the day, but you defo need to take care of yourself and try to get your voice back asap!

***

Free Text-to-Speech Apps

After yesterday's experience and while resting in bed, I decided to review a few other apps that may do the job nicely.

App 1: Talk

App 2: Type and Speak 

App 3: Easy Text to Speech

(I will review them together, as they are very similar and use the same voice database. The same text was read by both apps identically, and with the same voice. They only differ in their interface.)

• These apps allow you to paste your text, or type it straightaway, and play it out. (Easy Text to Speech does not require any text-pasting, it just reads from your clipboard)
• You are able to choose from a long list of languages and accents.
• You can export or save your audio file.
• You need an active data or Internet connection.
• Tip! You need to "overpunctuate" your text to avoid a robotic output.

App 4: IVONA Text-to-Speech

This application uses its own voice database. You choose an accent, and a speaker, and you download the voice database for that voice in particular. I have chosen "Amy, UK English". So far, it sounds like the most natural reading option, because the voice is somehow soothing. 

It has not, of course, reached the state of artificial intelligence, in that it does not chunk my text the way I would like it to, and it does not always employ the intonation I expect, but it does recognise a few chunks and it appears to be able to recognise patterns of declination and paratones. 

This application also allows you to set your preferred pitch and speech rate for the chosen voice.

From what I have been able to make out, it does not need an active Internet connection. The voice data files I have downloaded were 150 MB, so I guess it should all be "there".

App 4: Google Translate (Web version)

Even though it is not a text-to-speech app but rather a translator, I cannot help feeling surprised about the changes that Google has introduced to their voices and to the whole reading aloud experience. It still sounds pretty "robotic", but there are some chunks of language that the system automatically recognises and reads out with some fixed intonation patterns. This creates an overall "weird" effect, since you hear upsteps, downsteps and sudden tone changes that make the whole text sound incohesive at times, but less so, if compared to what it was like last year!

BTW! I have found a tip on how to get Google to read out a phrase out for you. Type in this address in your browser:

http://translate.google.com/translate_tts?tl=en&q=type your phrase here

And replace the bit that says "type your phrase here" with your own phrase, and voilá!

App 5: Bing Translator (Web version)

Much as I love Google, Microsoft has done a better job in terms of making their performance of read-out texts more natural. The Bing Translator option presents, in my humble opinion, a better chunking, and more natural selection of intonation patterns. 

See for yourself! Compare the output versions of a short paragraph of my lesson by both Google Translate and Bing Translator. Look out for some interesting word stress choices as well!

***

Final remarks and a few extras...

• A full selection of Android text-to-speech apps is available here.
• For a list of other possible uses for text-to-speech for language learning, and ideas to aid students with special needs, you can check out these resources:
• ‘Text-to-speech Technology’: What Does It Offer to Foreign Language Learners?http://callej.org/journal/7-2/Kilickaya.html

• Digital Support for English Language Teachers: http://blog.texthelp.com/2014/12/digital-supports-english-language-learners/
• Speech Recognition for Learning: http://www.brainline.org/content/2010/12/speech-recognition-for-learning_pageall.html

Hope this blog post has been of use! It has certainly inspired me to start considering writing a paper on synthetic-speech talk and intonation!

Podcast-y post #2: Your accent, your Patronus - Yet another reflection in passing

Some time ago, I recorded my first podcast on accents and the"Voldemort effect". Here's another post/ramble, unedited, heart-to-mind, mind-to-heart talk, of this complex feature of attaining, sustaining, and more importantly, owning and loving your accent.

It was inspired by my teacher trainees in their last year, and the challenges I find while teaching these advanced students.

Usual pre - listening warnings (I am a teacher, after all, and I am my own teacher, as well):
☂I have mispronounced "Patronus". I should have probably pronounced it: /pə'trəʊnəs/. ☂Plus, there are a few choices in word stress, nucleus placement, tone and tonality which I would have made differently. ☂A few " dysfluencies" as well, sorry to say!

As usual, many of these posts emerge as "reflections in passing", almost through a "stream of consciousness" technique, and editing this post would have meant spoiling the spontaneity of this moment of realisation, and thought.

So here it goes. Hope you will enjoy it.

Pronunciation, Phonesthesia and Realia: The case of the STRUT vowel (Part 2)

In my latest post, I compared General British STRUT and its variants to Riverplate Spanish /a/. I reviewed some articulatory and also acoustic features, and I also went over some basic notions of Speech Perception theories. I established that many of my Riverplate Spanish learners may associate English /ʌ/ with their Spanish /a/ "magnets" and may, thus, find the differentiation between both vowels challenging, both for perception and production.

(BTW, apparently, the differentiation between English /ʌ/ and other vowels appears to be quite an issue for many speakers of English as an L1 or L2, as Ettien Coffi (2014) reveals in this paper from the Proceedings of the 5th Pronunciation in Second Language Learning and Teaching Conference - page 11 onwards)

In this post, I would like to present some of the tips & tricks I have collected over the years (and I will try to acknowledge my colleagues' contributions whenever I can!). My claim will be that many "tips" do the job when they help create an image of the L2 sound which may help the learner steer away from their L1 quality, and that this may not necessarily just be reliant on making the right articulatory movements.

This is why some of my suggestions will draw on creating "mental images" through the contributions of Phonesthesia, and Realia; images which should appeal not only to visual aspects, but also auditory, and emotional resources.

***

A few definitions first:

Realia refers to the introduction of real-life objects in the classroom, to make the learning experience of certain concepts and routines more vivid. This technique also enables students to engage all their senses and learning styles. We can introduce actual objects, or we can create virtual situations that may allow students to experience similar emotions and actions as those recreated in the real situations.

(You can get some teaching ideas on using teaching aids in the TEFLSurvival blog, and the BusyTeacher website.)

Phonesthesia refers to the analysis of "sound symbolism". It basically studies how clusters of sounds may center around lexical sets that express similar meanings. So for example, in the Dictionary of Sound by Margaret Magnus, you can find a number of STRUT words that could be related to "puffy things":  plush, fuzz, fluff, cuff, muff, ruff.

This reminds me of a poem by Tony Mitton, called "Fluff"

What's this here?
A piece of fluff.
I don't know where I get this stuff.
I'll blow it away
with just one puff.
Huff!
There. That's enough.

So the combination of vowel /ʌ/ and the /f/ quality, reminiscent of blowing, creates this "puffy" effect of fluff and makes the poem lovely for oral performance, and effective! 

***

As a College student, I had a hard time fine-tuning my STRUT away from my Spanish /a/. I was given instructions, I  knew I had to drop my jaw, but still, it sounded pretty much like my own Spanish /a/. (Mind you, my friend and colleague Prof. Francisco Zabala has found that the STRUT quality as an allophone is present in many Spanish combinations of "a" + sound /x/, as in "caja".)  And I see my students at Teacher Training College producing a similar type of Spanish /a/ sound. So after a few tries, after watching native speakers of English produce their STRUT sounds, analysing the way this jaw-dropping takes place for this sound, I came up with the first articulatory tip that worked for some of my students:

"Keep to the railings of the mouth". I asked my students to imagine that each of the two sides of their lips, or the corner of their mouths, had a vertical railing, and that there should not be any smiling, as it would defy the railings of the mouth, and that the articulatory movement should be downwards, not sidewards. I could not help thinking of these special types of puppets ventriloquists use:

Image credit: David Noah. Flickr:https://www.flickr.com/photos/davidnoah/5170268541/

With this idea in mind, in one of my lessons, students were asked to place their fingers in the corners of their mouths in the puppet-like manner above, and look at themselves in a mirror/front facing camera of their mobile phones while producing STRUT words, trying to avoid a smile (which was tough, as they were having fun!).Therefore, this sound became the "puppet" sound (and for older Argentinians, this was the "Chirolita" sound, after a well-known ventriloquist in Argentina).

This articulatory tip did the trick for many students, but yet, not all of them really got to acquire a close quality; many students still produced a much fronter, or sometimes, opener vowel.

So I resorted to phonesthesia, and I thought of a few words I associated with the STRUT quality. By repeating the STRUT vowel to myself in isolation, I came up with these words (and a few others, after trying the marvellous Sound Search tool in the Cambridge Pronouncing Dictionary CD-Rom!)

• Things falling and making noise: thud, plunge  
• Unpredictable, shocking: abrupt, blunt, rough
• Pushing, stabbing(?): thrust, chuck, cut, nudge, punch, butt

(BTW (deviation alert!), as a Miranda Hart fan, I could help laughing at myself while repeating these words!)

So with some groups of students, I tried playing around with the words above, getting into the mood of things "falling" or happening "abruptely", which acts in much the same way the downward movement of the jaws does, like a small "bite", even.

Encouraged by the success of this tip for some students, I started thinking about the realia of "emotions", the "tone" that this sound evoked in me, and I could not help feeling "dull", "disgusted" or "miserable", as with these words:

• love
• money
• you suck!
• duh!
• yuck!
• f*#k

And I said to myself (and forgive the vulgarity of it all!), "what makes you suffer? Love, or money?". So I asked my students to think about their most "miserable" feeling, place themselves in that "sad place" (a bit like many actors do), "pull a miserable face" and go for /ʌ/ . 

Credits: http://i3.mirror.co.uk/incoming/article4311979.ece/ALTERNATES/s615b/Kim-Kardashian-and-Kanye-West-arrive-for-a-dinner-at-Hakkasan.jpg

(I have a confession to make: I sometimes ask my students to produce the f-word. It works! Perfect STRUT qualities EVERY time!)

When using this strategy a few years ago, a student came up with the memory of losing a match as a kid, and remembering his father's disappointed reaction (a very sad place to go to, if you ask me!). So we worked on that emotion, and we created a situation in which a very stern father or mother would push their son or daughter to win a race. And I came up with this terrible (!) poem/song, with a "run, run, run" chorus to the melody of Pink Floyd's "Run like Hell". It does sound like a bitter and very dark poem, but as a dramatic technique, it did the trick!

In one of the lessons where I tried the poem, students worked in two teams, with one group acting as an audience, singing the "run, run" Pink Floyd chorus, and a second group "mumbling" the words of the poem as they all pictured the race and their son running. The "little play" that resulted of this poem reminded me of Hermione in the audience uttering the Confundus charm and Harry saying "Come on, Ron" in a sort of low murmur (Deviation alert #2!):

***

Somehow this idea of a "miserable" STRUT sound has been the most effective tip to help my students provide a version of /ʌ/ that drove them closer to a quality different from their Spanish /a/, and also from the "Happy, cheerful, /æ/" and the "Relaxed, cute, awwww-like, /ɑ:/":

***

As with everything, you can find your own way of adapting these tricks for your own lessons, taking into consideration the "classroom management" factor, because let's face it, pronunciation work may create a bit of a mess in the classroom!

And of course, as Adrian Underhill claims, multisensory pronunciation learning is the key. Building one's own proprioception, and supplementing this with mental, auditory and emotional images should, in some way or other, contribute to the uniqueness of our learners' styles and processes. As I always say, what may work for one student may not necessarily work for the other! (Another piece of evidence for the "messy" nature of pronunciation work, sorry to say!)

***

Final thought: It is funny that many of the nice things of life may also take STRUT, and they don't seem to "fit" with with sound. What shall we do about 

love, fun, feeling chuffed, abundance, bubbles ...and maybe, money?

I leave that to you!

Pronunciation, Phonesthesia and Realia: The case of the STRUT vowel (part 1)

With the release of Adrian Underhill videos, I can now re-live some of the presentations of the IPA that my Phonetics tutor and trainer, Prof. Graciela Moyano, introduced when I was a first year student at College. It's just brilliant to go back to those moments of discovery of how our articulatory system works.

Adrian Underhill's phonemic chart. (I would probably do away with the centring diphthongs as they stand, using /ɛ:/ for /eə/ and maybe still keep /ɪə/)

I have completely embraced Adrian's belief and knowledge that "pronunciation is physical", and this motor side of pronunciation and our awareness of articulation through the different "buttons", as he calls them, is essential. However, out of his presentations, my independent reading, and my own teaching style, I have found that part of the process of acquisition of L2 features also lies on the images that we make of what the L2 features sound like.

Over the years, I have discovered that part of the fine-tuning processes we need to carry out to learn the sounds of L2 also involve affective aspects, including our own affective memory, and feelings. I have experienced situations in which students appear to be making the right articulatory movements, and still, there is something in the quality of the sound that does not "sound right". And in those cases, what I noticed is that it is the recourse to mental and affective images related to the sounds that appear to do the trick, rather than just the right lip or jaw position.

So in this post (which is, obviously, not a report on experimental research, so don't expect it to be so!), I will be going over a few techniques that I have used to include phonesthesia and realia in my pronunciation teaching. My focus this time will be on the STRUT vowel, /ʌ/, one which gave me a lot of trouble as a learner!

I will be covering a few different theoretical and practical areas on this post, which is why it will be divided into two:

Part 1: 

• Speech Perception theories: a review with links to further reading materials.
• English vowel STRUT /ʌ/and Spanish vowel /a/: a review.

Part 2:

• Tips and tricks to teach STRUT
• Articulation
• Realia
• Phonesthesia

***

Images, Categories, and Speech Perception Theories

When writing my research paper for my specialisation course in Phonetics I in 2006, I came across a huge number of acquisition and psycholinguistic theories which really helped me to make sense of and inform my teaching practices. Even though I am not an expert in acoustic or perceptual matters, I believe that some basic knowledge of theories of Speech Perception can actually aid our comprehension of what happens to our learners when faced with L2 pronunciation challenges.

Speech Perception Theories state that perception is categorical and not continuous (Liberman, 1985), that is, we perceive categories, no matter whether there is individual variation, and thus, we group sounds according to our perception. We can, in other words, say if a certain set of sounds are the same, or if they are different, despite the realisational differences of those sounds. The Perceptual Magnet Effect theory by Kuhl (1991) addresses similar issues for vowel sounds, by exploring in more detail how we can perceive the space or distance between the different variants we hear, to somehow establish what degrees of variation can be significant for us to associate a certain sound or stimulus to either another category, or to a "bad" (so to speak!) version of the category under scrutiny. This all appears to work pretty well for our L1, though there are a few "traps", as the McGurk effect reveals, when the information that the different senses provide do not appear to match:

If anything, the McGurk effect proves that we draw information from different multimodal sources, which is a very useful thing to know for those of us who train in L2 perception and production!

Now, when it comes to perception of L2 sounds, the "same-different" categorisation may fail, as we tend to interpret our L2 sounds from our own L1 targets, as Flege's Speech Learning Model (1995) states. So if we consider Kuhl's and Flege's models, we can predict that certain L2 qualities, if close to certain L1 targets, will be perceived as similar, and thus, learners may take longer to acquire them and create new categories for them.

Our learners, then, may create a mental image of what a particular L2 segment sounds like in the light of their already-existing images of the sounds in their L1. This means that in order to improve perception and production, we need to create new images and categories that will establish a perceptual distance between L1 and L2 sounds.

So in this post, I would like to explore the fact that one of the ways in which we can displace the L1 images in favour of L2 categories is by creating images that may exceed the information that the visual and tactile senses can give us for the articulation of the sound, and even the auditory information we can initially get. I would like to propose ways in which affective and visual memories can activate these new sounds in the shape of perceptual images.

***

The STRUT /ʌ/ vowel in General British

Cruttenden (2014:122) describes the STRUT vowel as a "centralised and slightly raised CV [a]" and he acknowledges the presence of a back-er quality [ʌ̞̈] in Conspicuous General British speakers, though this quality also appears to be heard more often in GB (I agree! See below). The jaws are said to be separated considerably, and the lips, neutrally open. There is an MRI of a phrase containing STRUT in Gimson's.... companion website here.

Collins and Mees (2012) have defined STRUT as a central, open-mid checked vowel, and admit a certain degree of variation, but especially in terms of fronter qualities. In much the same way, Mott (2011:117) describes STRUT as being "slightly forward of centre, just below half open, unrounded".

The abovementioned authors would then place STRUT somewhere around these areas:

Dark purple: STRUT, according to Mott & Collins and Mees. Light purple: Cruttenden (2014) and the backer quality [ʌ̞̈] he describes as a variant.

Geoff Lindsey has a great post called "STRUT for Dummies" with a myriad audio examples that illustrate his comments on the historical and present variation of STRUT, wavering between [ə] and [ɑ]-like qualities.

Many Argentinian colleagues and I believe that STRUT appears to be moving closer now to an [ɑ] quality, at least for many young speakers. I cannot claim to have the knowledge or "ear" that Lindsey happens to be blessed with, but here's an example of what I believe I hear pretty often:

Compare this young YouTuber Zoella's versions of "love", "until", "one", "somebody", "bump", "tummy", "just", "hundred". Apart from being great examples of intra-speaker variation (I personally don't hear all the versions of the word "love" with the same STRUT quality), there are some other interesting processes going on . (Also, a great taste of the kind of TRAP vowel I hear now, so different from General American versions!).

(BTW, according to Wikipedia, Zoella is from Wiltishire, and currently living in Brighton. What would you say her accent is?)

If we see these variations in acoustic terms (disclaimer: not at all my area of expertise...yet!), we can find in Rogers (2000) there is a table presenting formant values for RP vowels, based on a measurement of adult male speakers in Gimson, 1980. In that measurement, RP /ʌ/ has got the following values:
F1: 760 Hz
F2: 1320 Hz
( Confront with the values that the study by Wells in his 1962 study  had established F1:722, F2: 1236 )

(BTW, in simple terms, F1 inversely represents the articulatory tongue height, so that its higher frequency value represents lower tongue height. F2 represents degrees of backness, with lower values representing back-er vowels).

A study conducted by Hawkings and Midgley (2005) in four different age groups found the following mean formant values for these age groups:

• F1:630 -F2:1213 for over 65s (higher than the previous studies mentioned, and a bit backer)
• F1:643 F2:1215 for those speakers between 50-55 (lower than the over 65s, but equally central-back)
• F1:629 F2:1160 for speakers in the age range 35-40 (higher than those 10 years older, and backer)
• F1:668 F2:1208 for speakers aged 20-25 (lower than the other groups, but not as back as those speakers a bit older)

So according to this study, in comparison with the previous measurements reported, it would appear to be the case (at least for the speakers surveyed) that younger generations have a lower, and to a small degree, back-er STRUT than older generations).

If you are interested in pursuing this further, there are some interesting studies by
Ferragne and Pellegrini (2010), measuring vowels in 13 British accents.
Sidney Wood (SWPhonetics) analysis of different RP speakers' vowels across time.

***
Riverplate Spanish Vowel /a/

A contrastive analysis between General British and Riverplate Spanish (RS), and also the contributions of Speech Perception theories would lead us to expect that those students with RS as L1 are very likely to interpret English STRUT in the area of Spanish /a/ (not to mention distributional differences!). (BTW, in 2013 I attended a talk by Andrea Leceta at the III Jornadas de Fonética y Fonología, in which a comparative and experimental study on these features had been made, and the results confirmed these expectations. The proceedings have not yet been published, I am afraid.)

García Jurado and Arenas (2005) discuss Riverplate Spanish /a/ as having a wide degree pharyngeal of constriction (following the studies by Fant (1960) that focus the analysis of vowels based on the levels of constriction along the vocal tract) and clear oral opening and lack of lip rounding. They establish a mean F1:800 and F2:1200 values, which appear to match the degree of backness found for current versions of STRUT, though Spanish /a/ is much lower than English /ʌ/. 

Location of Spanish /a/ by Mott (2011)

You can see a cross-section of the articulation of /a/in Spanish and in English in these captures from the University of Iowa's Sounds of Speech app.

 Spanish /a/

(American) English /ʌ/


***
There appears to be an interesting similarity between Spanish /a/ and GB English /ʌ/ that lies mostly in the central part of the tongue employed, and an interesting difference, that is related to the height of the vowel and the level of jaw dropping, much higher than in Spanish, although they are both mid-to-low vowels. And still, it appears to me -and this is a very personal appreciation- that it is not jaw-dropping alone that makes our RS /a/ different from GB  /ʌ/. And my teaching of the sound over the years has also proven to me that jaw-dropping alone does not do the trick.

It may or it may not make sense to teach the exact quality of this sound to RS speakers, and it will depend on whether the focus of the lesson is on accentedness or on intelligibility, but the truth is that for an RS speaker, there needs to be a contrast between the three different qualities making up the contrast STRUT-TRAP-BATH in English, which to RS speakers may be only subsumed into one. 

So the next part of this post, coming up in a new weeks, I will present some "tips and tricks", rooted on some phonesthesic ideas, and also on realia techniques, that may help RS speakers turn their Spanish /a/s into STRUT.



After-post addition:
A reader rightly pointed out that I have not presented audio examples of my own of the difference between both sounds. So this is how I pronounce Spanish /a/ and English /ʌ/, and below you will find two Praat captures of the spectogram and formants of the same versions:

My Spanish /a/

My English /ʌ/