Listening Cherries is a blog where I talk about listening issues - from classroom activities to academic research. Listening Cherry no. 2 features a productive disagreement about a transcription.

A colleague visited me a short while ago, and we discussed a TED talk that I was working on. The TED talk is by Tom Wujec. I was excited (as I always am) about certain soundshapes that had occurred in his talk.

We listened to this extract (at 1:34 in the recording) in which I was interested in the three underlined words.

We disagreed on whether the word then had occurred. So the issue was, did Tom Wujec say 01 or 02 below?

[The underlined words are in a ‘squeeze zone’ - cf. Phonology for Listening, chapters 2 & 18, and incidentally - assuming that the word then does occur - this speech unit goes at 4.3 syllables per second, with the first three non-prominent syllables going at 9.2 sps and the two prominences going at 2.4 sps]

Here is the speech unit at half speed

Prior to hearing my colleague’s comments, I myself had had no doubts about the transcription - for me, Tom had said the three words and then they - and the TED transcript agreed with me. But I now have my doubts. And if I listen to the non-prominent section of this speech unit on its own, like this (you will hear it at both full and half speed), I hear something different:

Now I believe I hear |ən.eɪ| ‘ann-ey’ and not |ən.n̩.eɪ| ‘ann.en.ey’. So which is right? The three word version with then or the two-word version without then? For me, it is the three syllable version - because I assign priority to what I hear in the full speech unit.

But what I think is in some sense irrelevant, because this type of disagreement/uncertainty presents a wonderful learning/teaching opportunity. And it is essential not to insist that one transcription is right and that the other is wrong.

It is in the nature of spontaneous speech to be indeterminate - for stretches of the stream of speech to feature an in-between-y-ness, not definitely one thing or another. This is a point to be made that can be made through direct explanation in the classroom (e.g. ‘In the jungle, words resemble other words, and it is often difficult to determine with certainty which words have occurred’)

But for the purposes of teaching listening it is more important to get students to relish and savour the different ways in which these words might be said. For example, they can - following the instructions in the table below - savour the different versions of and then they by saying them aloud, as you (the teacher) cue the whole class. Then get them to work in pairs: student A prompts student B by selecting a row by saying ‘Row 1’, student B responds by saying and then they as appropriate for the row. After a minute, students change roles.

For more examples of and then they in TED talks, go here.

Listening Cherries is a blog where I talk about listening issues - from classroom activities to academic research. Listening Cherry no. 1 features a report from a Hong Kong classroom by James Pengelley.

One of the problems with the way teachers are trained to teach listening (in ELT) is that they are encouraged not to engage with the details of the sound substance of a recording (cf. my Grasping the nettle… here, and Phonology for Listening Chapter 16 here). We are told that research into listening in L1 shows that L1 listeners operate at the level of meaning, and do not listen to every word. The false conclusion is drawn that we should therefore encourage our students not to listen for every word, they should (after contextual preparation) listen for the stressed words, and seek to build meanings, and thereby understand. This is fine, but it ignores the learning opportunities offered by the recording as to how English sounds - it avoids the teaching and learning of the sound substance of the language.

Of course I would not want my students to inspect the soundshapes of every word in a three-minute recording. But it is worth spending a couple of minutes of class-time dipping in to a recording, and observing the sound shapes of words (they vary a huge amount). A colleague in Hong Kong, James Pengelley has alerted me to software available online, that makes it very easy to do this.

The software is mp3cut. You can move very quickly backwards and forwards through a recording to find an appropriate second or two to focus on, and to repeat it as often as you judge useful. As you will see from the website, it is easy to make ring tones for your iPhone from any recording. So (the thought occurred to me) if you have a class with many iPhone owners, you could set them the task of making a ringtone each from a particular recording - and get them to give it as a dictation to a fellow student.

James used it somewhat differently. Here’s what he said:

I had to cover a CAE class last night and was asked to do a listening practice with the group - we spent close to 2:15 hours on 3 practice questions and I used mp3cut.net to cut up the audios for them.

I first introduced them to the ‘and…and…and’ excerpt [sound file 8.2 from Phonology for Listening ] from the session on Wednesday and then we listened once and then spliced sections relating to the answers and then listened, repeated, spliced, counted the syllables, did some mini dictations….it was really interesting for them.

One of the students even commented that she felt listening in this way helped to her to understand why listening is so difficult for her, and where the mis-listenings come from and how some of the words change and how her brain reconstructs what she actually hears. It was a very interesting class - thought you’d appreciate the story 🙂