On this recording, fe016's volume vascillates quite a lot, as she moves her head. Most likely, she wore the headset around her neck for this one. Many examples, if anyone wants a sense of the variation in acoustic volume from this factor. Starts - No. No. No. That's a different thing. There's another - I don't know. It starts with a P_ or something. I forget the word for it, Oh. but it's - it's um Typically when you - you're ab- r- starting around forty for most people, it starts to harden and then it's just harder for the lens to Oh. shift things and th- the - the symptom is typically that you - Uh-huh. you have to hold stuff uh uh further away to - Yeah. to see it. In fact, uh m- my brother's a gerontological psychologist and he - he uh came up with an - an uh - a uh body age test which uh gets down to sort of only three measurements that are good enough st- statistical predictors of all the rest of it. And one of them is - is the distance Yeah. that you have to hold it at. Give someone a piece of paper and then they - Yeah. We're - we're live by the way, so we've got a good intro here Oh. Oh. Yeah. About how old I am. O_K. Yep. We can edit that out if you want. Oh, that's optional. No, that's O_K. Mm-hmm. O_K. So. You know . This time the form discussion should be very short, right? It also should be later. O_K. Because Jane uh is not here yet. Good point. And uh she'll be most interested in that. Uh, she's probably least involved in the signal-processing stuff so maybe we can just - just uh, I don't think we should go though an elaborate thing, but um uh Jose and I were just talking about the uh uh, speech e- energy thing, The @@ - Yeah. and I uh - We didn't talk about the derivatives. But I think, you know, the - the - i- if I can - if you don't mind my - my speaking for you for a bit, um Uh. Right now, that he's not really showing any kind of uh distinction, but uh - but we discussed a couple of the possible things that uh he can look at. Um. And uh one is that uh this is all in log energy and log energy is basically compressing the distances uh between things. Um Another is that he needs to play with the - the different uh uh temporal sizes. He was - he - he was taking everything over two hundred milliseconds uh, and uh he's going to vary that number and also look at moving windows, as we discussed before. Um And uh - and the other thing is that the - yeah doing the - subtracting off the mean and the variance in the - uh and dividing it by the standard deviation in the log domain, may not be the right thing to do. Hi. Yeah. Hi Jane! We just started. Could you take that mike there? Are these the long term means? Like, over the whole - I mean, the means of what? Thanks. Uh B- Between - All the frames in the conversation? Or of things that - between - No. Between - No. Neither. It's uh between the pauses uh for some segment. Oh. And so i- i- his - his - He's making the constraint it has to be at least two hundred milliseconds. Oh. And so you take that. And then he's - he's uh measuring at the frame level - still at the frame level, Right. of what - and then - and then just uh normalizing with that larger amount. um and - But one thing he was pointing out is when he - he looked at a bunch of examples in log domain, it is actually pretty hard to see the change. And you can sort of see that, because of j- of just putting it on the board that Yep. if you sort of have log-X_ plus log- X_ , Yeah, maybe it's not log distributed . that's the log of X_ plus the log of two and it's just, you know, it - it diminishes the effect of having two of them. Mmm. Yeah. @@ But you could do like a C_D_ F there instead? I mean, we don't know that the distribution here is Um. Yes, right. normally. So - So just some kind of a simple - So what I was suggesting to him is that - Actually, a P_D_F . But, you know, uh P_D_F But, either way. Yeah. Yeah. Yeah. Yeah, eith- eith- uh Something like that where it's sort of data driven. B- Yeah, but I think also u- I think a good first indicator is when the - the - the researcher looks at examples of the data and can not see a change in how big the - the signal is, Yeah. when the two speaker - Then, that's a problem right there. Yeah. Oh yeah. So. I think you should at least be able, doing casual looking and can get the sense, " Hey, there's something there. " and then you can play around with the measures. Oh yeah. Yeah. And when he's looking in the log domain he's not really seeing it. So. And when he's looking in straight energy he is, so that's a good place to start. Yeah. Um. So that was - that was the discussion we just had. Um. The other thing Actually we ca- had a question for Adam in this. How - Uh, when you did the sampling? uh over the speech segments or s- or sampling over the - the individual channels in order to do the e- uh the amplitude equalization, did you do it over just the entire - everything in the mike channels? You didn't try to find speech? No, I just took over the entire s- uh entire channel um sampled ten minutes randomly. Right, O_K. So then that means that someone who didn't speak very much would be largely represented by silence. Yep. And someone who would - who would be - So the normalization factor probably is i- i- i- Yeah, this was quite quick and dirty, and it was just for listening. is - is - Yeah. Yeah. O_K. And for listening it seems to work really well. Yeah. Yeah. Yeah. So. Yeah. But that's - Right. So th- But, it's not - Yeah. Not a good measure. O_K. So yeah there - there - there - There's a good chance then given that different people do talk different amounts that there is - Yeah. Yeah. there - there is still a lot more to be gained from gain norm- normalization with some Mmm. Yes, absolutely. sort if - if we can figure out a way to do it. Yeah. Uh. But we were agreed that in addition to that uh there should be s- stuff related to pitch and harmonics and so forth. Yeah. So we didn't talk at all about uh the other derivatives, but uh again just - just looking at - Uh, I think uh Liz has a very good point, that in fact it would be much more graphic just to show - Yeah. Well, actually, you do have some distributions here, Yeah. uh for these cases. You have some histograms, Yeah. um and uh, they don't look very separate. uh separated. This is the - the first derivate of log of frame energy What - Yeah. Yeah. uh without any kind of normalization. Yeah. Log energy. These the- Sorry. These are the - the first experiments uh with comment uh Frame energy. Except that it's hard to judge this because the - they're not normalized. It's just number of frames. Yeah. Yeah. Yeah. Yeah. But yeah, even so. W- I mean, what I meant is, even if you use linear, you know, raw "Number" - measures, like raw energy or whatever, maybe we shouldn't make any assumptions about the distribution's shape, and just use - you know, use the distribution to model the - the mean, or what- y- you know, rather than the mean take some - Yeah. But - And so in - in these he's got Yeah. that. He's got some pictures. Yeah. But he doesn't - he doesn't in the - he- i- just in derivatives, Yeah. but not in the - Oh. but he d- Right. So, we don't know what they look like on the, tsk - but he doesn't - doesn't - But he didn't h- have it for the energy. For the raw. He had it for the derivatives. Yeah. So. Yeah. I mean, there might be something there. I don't know. Huh. Yeah. Interesting Here I - I Oh that - yeah that's a good q- in - did - did you have this sort of thing, for just the - just the l- r- uh the - the unnormalized No log energy? I - I - I haven't the result O_K. Yeah. So she - she's right. That's a - but it's the - it's the - the - the following. Well it might be just good to know what it looks like. Cuz - Yeah. That's - Huh? That's uh cuz I'd mentioned scatter plots before but she's right, I mean, even before you get the scatter plots, just looking at a single feature Yeah. uh, looking at the distribution, is a good thing to do. Catal- uh - Combining the different possibilities of uh the parameters. I - I - I - I mean the - the - the scatter plot combining eh different n- two combination. combination of two, of energy and derivate - Yeah, but - but what she's saying is, which is right, is le- I mean, let's start with the - Before we get complicated, let's start with the most basic wh- thing, which is we're arguing that if you take energy - uh if you look at the energy, that, when two people are speaking at the same time, usually there'll be more energy than when one is right? Yeah. That's - that sort of hypothesis. And the first way you'd look at that, That's right. uh s- she's, you know, absolutely right, is that you would just take a look at the distribution of those two things, Yeah. much as you've plotted them here, You know, but just - but just - Yeah. just uh do it - Well in this case you have three. You have the silence, and that - that's fine. Yeah. So, uh with three colors or three shades or whatever, just - just look at those distributions. Yeah. And then, given that as a base, you can see if that gets improved, you know, or - or - or worsened by the - looking at regular energy, looking at log energy, we were just proposing that maybe it's - you know, it's harder to see with the log energy, Yeah. um and uh also these different normalizations, does a particular choice of normalization make it better? But I had maybe made it too complicated by suggesting early on, that you look at scatter plots because that's looking at a distribution in two dimensions. Yeah. Let's start off just in one, Yeah. uh, with this feature. Yeah. I think that's probably the most basic thing, before anything very complicated. Yeah. Yeah. Um And then we- w- I think we're agreed that I agree, yeah. pitch-related things are - are - are going to be a - a really likely candidate Uh-huh. to help. Um O_K. But since - uh your intuition from looking at some of the data, is that when you looked at the regular energy, that it did in fact usually go up, when two people were talking, Yeah. Yeah. Yeah, yeah, yeah. that's - eh you know, you should be able to come up with a measure which will match your intuition. And she's right, that a - that having a - But - having - having this table, with a whole bunch of things, with the standard deviation, the variance and so forth, Uh-huh. it's - it's - it's harder to interpret than just looking at the - Yeah. But - the same kind of picture you have here. It - it's curious but uh I f- I found it in the - in the mixed file, in one channel that eh in several - oh e- eh several times eh you have an speaker talking alone Mm-hmm. with a high level of energy Mm-hmm. eh in the middle eh a zone of overlapping Mm-hmm. with mmm less energy and eh come with another speaker with high energy and the Mm-hmm. overlapping zone has eh less energy. Yeah. So there'll be some cases for which - Because there reach very many @@ Right. But, the qu- So - So they'll be - This is - I w- want to point to visual things, But I mean they - there'll be time - There'll be overlap between the distributions, but the question is, Yeah. "If it's a reasonable feature at all, there's some separation." Yeah. Especially locally. So. Locally. And the other thing is I- just locally , yeah. Mm-hmm. And - Sorry. I - I was just going to say that - that right now we're just exploring. What you would imagine eventually, is that you'll feed all of these features into some Yeah. discriminative system. Yeah. Yeah. And so even if - if one of the features does a good job at one type of overlap, another feature might do a good job at another type of overlap. Yeah. Yeah. Yeah, this is the - Right. I mean the - the reason I had suggested the scatter f- p- features is I used to do this a lot, when we had thirteen or fifteen or twenty features to look at. um Because something is a good feature uh by itself, you don't really know how it'll behave in combination and so it's nice to have as many - as many together at the same time as possible in uh in some reasonable visual form. There's cool graphic things people have had sometimes to put together three or four in some funny - funny way. But it's true that you shouldn't do any of that unless you know that the individual ones, at least, have - Yeah. have some uh - Well, especially for normalizing. I mean, some hope Mm-hmm. it's really important to pick a normalization that matches the distribution for that feature. Mm-hmm. And it may not be the same for all the types of overlaps or the windows may not be the same. e- Actually, I was wondering, right now you're taking a - all of the speech, from the whole meeting, and you're trying to find points of overlap, but we don't really know which speaker is overlapping with which speaker, right? Right. Yeah. So I mean another way would just be to take the speech from just, say, Morgan, And just Jane and then just their overlaps, like - but by hand, by cheating, and looking at you know, if you can detect something that way, because if we can't do it that way, there's no No prayer. good way that we're going to be able to do it. That - You know, there might be something helpful and cleaner about looking at just individuals and then that combination alone. Yeah. Plus, I think it Mm-hmm. has more elegant - e- The m- the right model will be easier to see that way. So if - I don't know, if you go through and you find Yeah. Adam, cuz he has a lot of overlaps and some other speaker who also has e- enough speech and just sort of look at those Yeah. three cases of Adam and the other person and the overlaps, maybe - and just look at the distributions, maybe there is a clear Yeah. Uh-huh. pattern but we just can't see it because there's too many combinations of - of people Yeah. that can overlap. I had the same intuition last - last - last week. So. Yeah. Just seems sort of complex. I think it's - to start with it's s- your - your idea of simplifying, starting with something that you can see eh you know without the extra layers of - Right. Cuz if energy doesn't matter there, like - I don't think this is true, but what if To study individual? Sorry, what? Hmm? To study individual? Well, you - you - you don't have to study everybody individually but The - the - the - Well, to study the simplest case But - just simple case and to get rid of extra - Consider - the one that has the lot of data associated with it. Right. Cuz what if it's the case and I don't think this is true - That was a great overlap by the way. What if it's the case that when two people overlap they equate their - you know, there's a conservation of energy and everybody - both people talk more softly? I don't think this happens at all. Or - or what if what if the equipment - Or they get louder. what if the equipment adjusts somehow, there's some equalizing in there? Yeah or - Uh, I mean. no we don't have that. But. Well, but - O_K. But I think that's what I was saying about different types of Saturation. overlap. There are - there are different types, and within those types, like as Jose was saying, that sounded like a backchannel overlap, meaning the kind that's a friendly encouragement, like "Mm-hmm.", "Great!", "Yeah!" Yeah. And it doesn't take - you don't take the floor. Um, but, some of those, as you showed, I think can be discriminated by the duration of the overlap. Yeah. So. It - Actually the s- new student, Don, who um Adam has met, and he was at one of our meetings - He's getting his feet wet and then he'll be starting again in mid-January. He's interested in trying to distinguish the types of overlap. I don't know if he's talked with you yet. Yeah. But in sort of honing in on these different types and - I don't consi- Now I don't consider that possibility. This is a s- a general studio of the overlapping So maybe - Yeah. we're studying the - So it might be something that we can Well I - help by categorizing some of them and then, i- I - I - I would s- you know, look at that. actually still recommend that he do the overall thing because it would be the quickest thing for him to do. He could - You see, he already has all his stuff in place, he has the histogram mechanism, Yeah. he has the stuff that subtracts out - and all he has to do is change it uh uh from - from log to plain energy and plot the histogram and look at it. And then he should go on and do the other stuff Yeah. bec- but - Yeah, no. I didn't mean that - But this will - that - for you to do that, but I was thinking if - if Don and I are trying to get categories and we label some data for you, and we say this is what we think is going - Mm-hmm. Mm-hmm. So you don't have to worry about it. And here's the three types of overlaps. And we'll - we'll do the labeling for you. Yeah. Yeah. Hm- hmm. Um. Consider different class of overlap? Yeah, that we would be working on anyway. If there's time. Yeah. Then maybe you can try some different things for those three cases, and see if that helps, or - Yeah. This is the thing I - I comment with you before, that uh we have a great variation of th- situation of overlapping. Mm-hmm. And the behavior for energy is, uh log energy, is not uh the same all the time. Mm-hmm. And - But I guess I was just saying that - that right now uh from the means that you gave, I don't have any sense of whether even, you know, there are any significant number of cases for which there is distinct - and I would imagine there should be Yeah. some - you know, there should be - The distributions should be somewhat separated. Yeah. Uh and I - I would still guess that if they are not separated at all, Yeah. that there's some - there's - there's most likely something wrong in the way that we're measuring it. Yeah. Yeah. Um, but um For instance, I mean I wouldn't expect that it was very common overall, that when two people were talking at the same time, that it would - that it really was lower, although sometimes, as you say, it would. Yeah. Yeah. Yeah, no, that was - That was a jok- So. So. or a sort of, Yeah. a case where - where you would never know that unless you actually I mean. No. It could - it probably does happen sometimes. Yeah. go and look at Right. Yeah. Yeah. two individuals. So. Yeah. Mind if I turned that light off? The flickering is annoying me. O_K. It might the case, though, that the significant energy, just as Jose was saying, comes in the non-backchannel cases. Because in back- Most people when they're talking don't change their own energy when they get a backchannel, cuz they're not really predicting the backchannel. Mm-hmm. And sometimes it's a nod and sometimes it's an "mm-hmm". And the "mm-hmm " is really usually very low energy. Yeah. Yeah. So maybe those don't actually have much difference in energy. But all the other cases might. e- and the backchannels are sort of easy to spot s- in terms of their words or - I mean, just listen to it. So. Yeah. e- But - and - and again what they - what difference there was would kind of be lost in taking the log, so, Well, it would be lost no matter what you do. It just - as well. Yeah. Mmm, no, if it's - if i- if it's - Tone Well, it won't be as big. I mean, even if you take the log, you can - your model just has a more sensitive measures. So. Yeah. Sure, but tone might be very Yeah, you're "mm-hmm" tone is going to be very different. Yeah. Right. Right. You could imagine doing specialized ones for different types of backchannels, if you could - if you had a good model for it. Your "mm-hmm" detector. If - if you're - a- I guess my point is, if you're doing essentially a linear separation, taking the log first does in fact make it harder to separate. Right. Yeah. So it's - So, uh if you i- i- So i- if there - if there close to things it does Yeah. it's a nonlinear operation that does in fact change the distinction. If you're doing a non- if you're doing some fancy thing then - then yeah. And right now we're essentially doing this linear thing by looking across here and - and saying we're going to cut it here. Um and that - that's the indicator that we're getting. But anyway, yeah, we're not disagreeing on any of this, we should look at it more uh - more finely, but uh uh I think that - This often happens, you do fairly complicated things, and then you stand back from them and you realize that you haven't done something simple. So uh, if you generated something like that just for the energy and see, and then, a- a- a- as - as Liz says, when they g- have uh uh smaller um, more coherent groups to look at, that would be another interesting thing later. Uh-huh. Uh-huh. And then that should give us some indication - between those, should give us some indication of whether there's anything to be achieved f- from energy at all. And then you can move on to the uh uh more pitch related stuff. Mm-hmm. I - I - I think this is a good idea. O_K. Not consider the log Mm-hmm. energy. Yeah. @@ But then the - Have you started looking at the pitch related stuff at all, or - ? The - ? Pitch related? Harmonicity and so on? I - I'm preparing the - the program but I don't - I don't begin Preparing to - because eh Yeah. I saw your email and I agree with you Yeah. it's better to - I suppose it's better to - to consider the - the energy this kind of parameter Oh, that's not what I meant. bef- No, no. I - I - I - I - Well, we certainly should see this but I - I - I - I think that the harm- I certainly wasn't saying this was better than the harmonicity and pitch related things I was just saying I - I go on with the - with the Yeah. pitch, aha! O_K. Yeah, I was just saying - I - I - I - I understood uh that eh I - I had to finish by the moment with the and - and concentrate my - my energy in that problem. O_K. O_K. O_K. But I think, like, all these derivatives and second derivatives and all these other very fancy things, I think I would just sort of look at the energy and then get into the harmonicity as - as O_K. I go on a suggestion. with the pitch. Uh O_K. So maybe uh since w- we're trying to uh compress the meeting, um, I know Adam had some form stuff he wanted to talk about and did you have some? I wanted to ask just s- something on the end of this top- topic. So, when I presented my results about the uh distribution of overlaps and the speakers and the profiles of the speakers, Uh-huh. at the bottom of that I did have a proposal, and I had plan to go through with it, of - of co- coding the types of overlaps that people were involved in s- just with reference to speaker style so, Oh. you know, with reference - and you know I said that on my - in my summary, that That'd be great. Yeah, I remem- Right. you know so it's like people may have different amounts of being overlapped with or overlapping but that in itself is not informative without knowing what types of overlaps they're involved in so I was planning to do a taxonomy of types overlaps with reference to that. That would be great. That would be really great. Yeah. So, but it you know it's like Hmm. it sounds like you also have uh something in that direction. Is - is it - We have nothing - You know, basically, we got his environment set up. He's - he's a double-E_ you know. So. It's mostly that, if we had to label it ourselves, we - we would or we'd have to, to get started, but if - It - it would be much better if you can do it. You'd be much better at doing it also because you know, I - I'm not - Interesting. I don't have a good feel for how they should be sorted out, and I really didn't wanna go into that if I didn't have to. So if - If you're w- willing to do that or - or - Well maybe we can O_K. It would be interesting, though, to talk, maybe not at the meeting, Yeah. but at some other time about what are the classes. Mm-hmm. Yeah. Yeah. Mm-hmm. I think that's a research effort in and of itself, Yeah, it would be interesting. because you can read the literature, but I don't know how it'll turn out and, Yeah. It seems like we also s- with reference to a purpose, too, that we we'd want to have them coded. I would think it's interesting, yeah. You know, it's always an interesting question. Yeah. That'd be great. Yeah. Yep. That'd be really great. And we'd still have some funding O_K. I can do that. uh- uh- for this project, like probably, if we had to hire some - like an undergrad, because uh Don is being hhh. covered half time on something else - Mm-hmm. I mean, he - we're not paying him the full R_A-ship for - all the time. So. um If we got it to where we wanted - we needed someone to do that - I don't think there's really enough data where - where - Mm-hmm. Yeah, I see this as a prototype, Yeah. to use the only the - Yeah. the already transcribed meeting as just a prototype. I - I think a- a- another parameter we c- we - But - we can consider is eh the duration. Mm-hmm. Another e- e- m- besides eh the - the class of overlap, the duration. Because is possible eh some s- s- um eh some classes eh has eh a type of a duration, eh, a duration very short Mm-hmm. uh when we have - Yeah, definitely. Yeah, maybe - It may be correlated. we have overlapping with speech. Is possible Mm-hmm. to have. And it's interesting, I think, to consider the - the window of normalization, normalization window. Eh because eh if we have a type of, a kind of eh overlap, eh backchannel overlap, with a short duration, is possible eh to normali- i- i- that if we normalize eh with eh eh consider only the - the eh window eh by the left eh ri- eh side on the right side overlapping with a - a very eh oh a small window eh the - if the fit of normalization is eh mmm bigger eh in that overlapping zone eh very short Mm-hmm. Yeah, that's true. The window shouldn't be larger than the backchannel. I - I me- I - I understand. I mean that you have Yeah. eh you have a backchannel, eh, eh - you have a overlapping zone very short and you consider eh n- eh all the channel to normalize Mm-hmm. this very short eh - Mm-hmm. for example "mmm mm-hmm hmm" eh And the energy is not eh height eh I think if you consider all the channel to normalize and the channel is mmm bigger Mm-hmm. eh eh eh compared with the - with the overlapping eh duration, eh the effect is mmm stronger eh that I - I mean the - the e- effect of the normalization eh with the mean and the - and the variance Mm-hmm. eh is different that if you consider only a window compared eh with the n- the duration of overlapping. Not - You - you want it around the overlapping part. You want it to include something that's not in overlapping but - Yeah. Yeah. Yeah. Mm-hmm. but uh I - I don't know. Is - s- Yeah. Well it's a sliding window, right? So if you take the - the measure If - Mm-hmm. Yeah. in the center of the overlapped piece, Yeah. you know, there'd better be some- something. But if your window is really huge then yeah you're right you won't even - The portion of the - Yeah, This is the - of the backchannel won't - won't effect anything. But you - This is the - the idea, to consider only the - the small window Yeah. So. near - near - near the - the overlapping zone. You know, you shouldn't be more than like - You should definitely not be three times as big as your - as your backchannel. Yeah. Mm-hmm. Then you're gonna w- have a wash. And hopefully it's more like I'm not sure that's necessarily true. on the order of - Yeah? It is an empirical question, it seems like. Yea- Because - Yeah. because it - because Yeah. um again if you're just compensating for the gain, you know, the fact that this - this gain thing was crude, and the gain wh- if someone is speaking relatively at consistent level, just to - to give a - an extreme example, all you're doing is compensating for that. And then you still s- And then if you look at the frame with respect to that, it still should - should uh change Yeah, it depends how different your normalization is, as you slide your window across. I mean. That's something we don't know. Mm-hmm. It's possible to try it both ways, isn't it? Well, I mean we're also talking about in this small @@ a couple of different things. I mean, one is your analysis window and then the other is any sort of normalization that you're doing. Yeah I was talking about the n- normalization window. And the - And they could be quite different. Yeah. Yeah. Right. This was sort of where - Yeah. That's true. Yep. where we were last week. Yeah. But, anyway We - we'll have to look at some core things. Um. O_K. O_K. But that'd be great if - if you're marking those and - Great. O_K. um. Yeah. But it is definitely true that we need to have the time marks, Mm-hmm. and I was assuming that will be inherited because, Yep, I agree. if you have the words and they're roughly aligned in time Mm-hmm. via forced alignment or whatever we end up using, then you know, this student and I would be looking at the time marks and classifying all the frames inside those as whatever labels Coming off of the other - Yeah. Good. So, it wouldn't be Jane gave I wasn't planning to label the time marks. I can give you I was thinking that that would come from the engineering side, yeah. my transcription file, no? @@ I don't think you need to. Yeah. That should be linked to the words which are linked to time somehow, right? There you go. Well we're not any time soon going to get a forced alignment. So. Not now. Yeah. Um If it's not hand-marked then we're not going to get the times. Well, it's something that w- Well, we - we wouldn't be able to do any work without a forced alignment anyway, so somehow if - Yes once he gets going we're gonna hafta come up with one and Yes. Yeah. I mean w- I guess we could do a very bad one with Broadcast News. Good. So whatever you would label would be Good. attached to the words, I think. Great! Good, good. Mm-hmm. Yeah. Well again for the close mike stuff, That might be good enough. we could come up - take a s- take the Switchboard system or something, and - Yeah. Um It'd be worth a try. It would be interesting to see what we get. Just, you know, low-pass filter the speech and - Cuz there's - there's a lot of work you can't do without that, I mean, how - how would you - Yeah. You'd have to go in and measure every start and stop point next to a word is y- if you're interested in anything to do with words. So. Yep. It would be very inefficient. Yeah. Mm-hmm. Anyway So that'd be great. Good. O_K. Yeah. There's something we should talk about later but maybe not just now. But, uh, should talk about our options as far as the uh uh transcription Yep, if I_B_M doesn't - But. Well, w- O_K. Good. But Do we hafta turn - we'll do that later. Are we supposed to keep recording here? Yeah. Let's do that later. Yeah Right. Yeah. We'll talk about it later. Yeah. So uh Uh "forms". You had something on forms. Forms Next iteration of forms. Oops. Oh! Oh good, O_K. Um. Oh. How - So it's two pages per person? Nope. One's a digit form, one's a speaker form. Oh! So one is a one time only speaker form and the other is the digits. Oh, I see. Oh it's the same. Oh no no. Is - is new Is O_K. So don't fill these out. Alright. This is just the suggestion for uh what the new forms would look like. So, they incorporate the changes that we talked about. Date and time. Uh why did you switch the order of the Date and Time fields? This is rather a low-level, but On which one? On - on the new one, Time comes first and then Date, but I thought - Oh you mean on the digit form? This is - this is rather a low level question, but - Uh, because but it used - used to be Date came first. the user fills out the first three fields and I fill out the rest. Oh I see. Well, how would the - So it was intentional. It's an interesting observation, but it was intentional. How would the user know the time if they didn't know the date? Because the date is when you actually read the digits and the time and, excuse me, the time is when you actually read the digits, but I'm filling out the date beforehand. If you look at the form in front of you? that you're going to fill out when you read the digits? Yeah. you'll see I've already filled in the date but not the time. I always assumed - So the time is supposed to be pretty exact, because I've just been taking beginning time - Yeah, me too. time of the meeting. Yeah, Yeah, I've noticed that in the forms. I - yeah. The - the reason I put the time in, is so that the person @@ Me too. who's extracting the digits, meaning me, Oh! will know where to look in the meeting, Oh dear. We've been - But - to try to find the digits. we've been messing up your forms. I know. I am put - I am putting the beginning of the meeting. So you should call it, like, "digits start time". Or. And I haven't said anything. Yep. in - on there . Why - What - what were you putting in? Oh, well, I was saying if we started the meeting at Yeah. two thirty, I'd put two thirty, and I guess d- e- everyone was putting two thirty, and Yeah. Oh. No, it's about fifty fifty. I didn't realize there was "uh oh I'm about to read this and I should" - Actually it's about one third each. About one third of them are blank, about one third of them are when the digits are read, and about one third of them are when the meeting starts. Oh. So. This would be a radical suggestion but - I could put instructions? Nah. Ei- either that or maybe you could Yeah. maybe write down when people start reading digits on that particular session. But if I'm not at the meeting, I can't do that. I know, O_K. Yeah, he's been setting stuff up and going away. So. That's a good point. I see. Good point good point. For some reason he doesn't want to sit through every meeting that's - Yep, but that is the reason Name, Email and Time are where they are. Oh, O_K. Alright. Yeah. I rest my - And then the others are later on. Uh-huh. O_K. w- And the Seat is this number? Mm-hmm. Seat and Session. "For official use only" That's - "use only" Well, he's very professional. Actually you could - Well that does raise another question, which is why is the "Professional use only" line not higher? Why doesn't it come in at the point of Date and Seat? Oh. What? Because we're filling in other things. What? Well, because - If y- your - your professional use, you're gonna already have the date, and the s- What - which form are you talking about? Well I'm comparing the new one with the old one. This is the digit form. Oh. Oh you're talking about the digit form. The digit form doesn't - Yeah. Digit. Digit form. Oh! I wasn't supposed to - Yeah. Sorry. The digit - No, that's alright. The digit form doesn't have a "for official use only" line. Sorry. It just has a line, which is what you're supposed to read. That - uh O_K. @@ So on the digits form, Sorry about that. Yeah. everything above the line is a fill-in form and everything below the line is digits that the user reads. Yeah. O_K. Alright s- but I didn't mean to derail our discussion here, so you really wanted to start with this other form. No, either way is fine I just - You just started talking about something, and I didn't know which form you were referring to. Alright yeah, I was comparing - so th- this is - So I was looking at the change first. So it's like we started with this and now we've got a new version of it wi- with reference to this. So the digit form, we had one already. Now the f- the fields are slightly different. So the main thing that the person fills out um is the name and email and Yeah. time? Right. Ah! You do the rest? Yep. Just as uh - as I have Right. for all the others. What - And there's an addition of the native language, which is a bit redundant. This one has Native Language and this one does too. That's because the one, the digit form that has native language is the old form not the new form. Oh! Thank you. Thank you, thank you. There we go. Yeah. Oh, yeah. I'll catch up here. O_K, I see. "South Midland, North Midland" That's the old and that's the new. Yeah this was the problem with these categories, I - I picked those categories from TIMIT. I don't know what those are. Actually, the only way I know is from working with the database and having to figure it out. What - With TIMIT, yeah? uh-huh. What i- So is South Midland like Kansas? and North Midland like - like uh Illinois, or - ? So, I was gonna ask wh- w- Well yeah. Yeah. Nor- um - I mean. So - so what accent are we speaking? Western? By definition? And for simple for - for me? Is mean my native language Spanish - Probably Western, yeah. Well, Spanish? eh The original is the center of Spain and the beca- Yeah, I mean you could call it whatever you want. For the foreign language we couldn't classify every single one. So I just left it blank and you can put whatever you want. Because is different, the Span- uh the Spanish language from the - the north of Spain, of the south, of the west and the - Sure. But. So I'm not sure what to do about the Region field for English variety. Yeah. You know, when I wrote - I was writing those down, I was thinking, "You know, these are great if you're a linguist ". Yeah. But I don't know how to - I don't know how to - I don't know how to categorize them. Actually even if you Yeah. t- If you're - if e- This wasn't developed by - th- these regions weren't - if y- if you're a T_I or M_I_T from nineteen eighty-five. Yeah Yeah. So I guess my only question was if - if you were a South Midland speaking region, person? Mm-hmm. Would you know it? Is that what you would call yourself? I don't know. Yeah. You know, I think if you're talking - if you're thinking in terms of places, as opposed to names different peop- names people have given to different ways of talking, I would think North Midwest, and South Midwest would be more common than saying Midland, right, I mean, I - Yeah. Now the usage - I went to s- Maybe we can give them a li- like a little map? with the regions and they just - No, I'm serious. No, that's not bad. Because i- at this - Yeah. it takes less time, and it's sort of cute there's no figure. in that side - in that side of the - the paper. Well. Well just a little - You know, it doesn't have all the detail, but you sort of - But what if you moved five times and - and uh Well, I was thinking you could have ma- multiple ones No, but you're categorized. That's the same - and then the amount of time - so, roughly. So. You could say, you know "ten years on the east coast, Well, five years on the west coast" or something or other. We - I think we don't want to get that level of detail at this form. I think that's alright if we want to follow up. But. I guess we don't really know. I mean I - As I said, I don't think there's a huge benefit to this region thing. It - it gets - The problem is that for some things it's really clear and usually listening to it you can tell right away if it's a New York or Boston accent, but New York and Boston are two - well, I guess they have the N_Y_C, but New England has a bunch of very different dialects and - Mm-hmm. and so does um Yeah, so I picked these regions cuz we had talked about TIMIT, and those are right from TIMIT. S- So do other places. Right. And so these would be satisfying like So. a speech research community if we released the database, but as to whether subjects know where they're from, I'm not sure because um I know that they had to fill this out for Switchboard. This is i- almost exactly the same as Switchboard regions or Oh. O_K. very close. Yeah. Um And I don't know how they filled that out. But th- if Midland - Yeah, Midland is the one that's difficult I guess. I think a lot of people - Yeah. Also Northwest you've got Oreg- Washington and Oregon now which uh y- people don't know if it's western or northern. Yeah, I certainly don't. I mean, I was saying I don't even know what I speak. It's like Northwest Oh, what is Northern? Am I speaking - Am I speaking Western? Well and what - and what's Northern? I think originally it was North - Northwest Northwest? Yeah. But - Yeah, so this is a real problem. I don't know what to do about it. Yeah. I wouldn't know how to characterize mine either. And - and so I would think - I c- I would say, I've - I've got a mix of California and Ohio. I think at the first level, for example, we speak the same. I don't know. our - our dialects Uh-huh. Or whatever you - region are the same. But I don't know what it is. So. Well, you have a like techno-speak accent I think. a techno-speak accent? A techno- Yeah, you know? A - a geek region? Well it's - Geek region. I mean I - you can sort of identify it f- It's - it's - Is different. Is different. not - not that that's - but - but maybe that - maybe we could leave this and see what people - See what people choose and then um let them just fill in if they don't - I mean I don't know what else we can do, cuz - I'm wondering about a question like, "Where are you from mostly?" Yeah. But I - I'm s- Yeah. I'm - now that you mentioned it though, I am - really am confused by "Northern". I agree. I really am. I mean, I agree. Yeah. I agree. if - if you're in New England, that's North. If you're - i- if you're Scandinavian, the Minnesota area's north. Uh yeah. That's - That's North Midland. But that's also North Midland, right? Yeah. Oh, @@ . - O_K. And - and - and Oregon and - Yeah. Of course, that's very different from, like, Michigan, and Oregon and Washington are - are Western, but Mmm. or - they're also Northern. Mm-hmm. uh, Idaho? Well there are hardly any subjects from Idaho. Montana? No problem. Just rule them out. There's only a few people in Idaho. There are hardly any subjects from "beep" Yeah. Sorry. No, that's - And - is - in those - Maybe - Maybe we - Maybe we should put a little map and say "put an X_ on where you're from", Yeah really. And if you put - We could ask where they're from. Yeah. But- It'd be pretty simple, yeah. Yeah. We went back to that. If you put eh the state? Well well we sort of - Where are you from mostly? We - we went - we went around this and then a lot of people Uh-huh. ended up saying that it - Mm-hmm. You know. Well, I like the idea of asking "what variety of English do you speak" as opposed to where you're from Yeah. Because th- if we start asking where we're from, again you have to start saying, "well, is that the language you speak or is that just where you're from? " Hmm? Right. Right. Yeah. I mean it gives us good information on where they're from, but Let's - Mm-hmm. And - that doesn't tell us anything - well, enough about their - We could always ask them if they're from - I mean. So - so I would say Germany like - You know am I speaking with German accent I don't think so. Oh. Right. Well, see, I'm thinking "Where are you from mostly " Oh, O_K yeah. because, you know, then you have some - some kind of subjective amount of time factored into it. Yeah. Yeah. Yep. Yeah. Yeah, I guess I could try to put - squeeze in a little map. I mean there's not a lot of r- of room I'd say, uh, "Boston, New York City, the South and Regular". Well - I think of those, Northern is the only one that I don't even know what they're meaning. And - And - Oh, I don't know. Yeah. Yeah. That's a joke. That's - So let's make it up. S- I mean, who cares. Right? We can make up our own - So we can say "Northwest", "Rest of West" or something. You know. "West" and Ye- hhh. I don't think the Northwest people speak any differently than I do. Um I mean. It doesn't even - Yeah, exactly. That's not really a region. "Do you come from the Louisiana Purchase?" I - So we could take out "North" - And usually here - people here know what is their "Northern". @@ kind of mmm lang- English language? That - that's exactly what we're arguing about. We don't know. eh here- That's - Yeah, w- It's - In - Is easy for people to know? It's - it's harder in America anywhere else, basically. because you have - I mean some of them are very obvious. If you - if you talk to someone speaking with Southern drawl, you know. N- m- Yeah. Yeah, or Boston. Or Boston, yeah. I can't do it, but - Or Boston? And those people, if you ask them to self-identify their accent they know. Yeah. Yeah. Yeah, they do. They know very well. Yeah I agree I agree. I agree. They know they don't speak the same as the But And they're proud of it. day o- Yeah. Yeah, exactly. is Boston New England? It's identity thing. And they're glad to tell you. @@ style. Well. Depends who you ask, I suppose. W- Oh. I guess that's the problem with these categories. But that's why they have New York City but - Well, we ca- Well, Boston's @@ , too . Well, why can't we just say characterize - Or - something like char- characterize your accent "Characterize your accent if you can." and - and so I would say, "I don't know". Yeah. Right, which probably means you have a very - But someone from Boston with a really strong coloration would know. And so would an R_less Maine - And that's actually good. Yeah. or something, yeah. I was - I was thinking of something along that line because if you don't know, How- Good. then, you know, ruling out the fact that you're totally inept or something, Hmm. if somebody doesn't know, it probably means their accent isn't very strong compared to the sort of Well, I mean, midwest Hmm? it wasn't that long ago that we had somebody here who was from Texas who was absolutely sure that he didn't have any accent left. standard. And - and had - he had a pretty O_K, so. I propose, take out Northern noticeable drawl. add, don't know. Oh. Yeah. I - I would say more - more sweepingly, "how would you characterize your Yeah. accent?" So you want to change the instructions also not just say region? W- Well, I think this discussion has made me think that's s- something to consider. I don't know if I - if I read this form, I think they're going to ask it - they're going to answer the same way if you say, "What's variety of English do you speak? Region." as if you say "what variety of region - region do you speak? Please characterize your accent?" They're going to answer the same way. I guess - Well, I was not sure that - I - Mmm. So. I was suggesting not having the options, just having them - Oh, I see. Huh. Well what we talked about with that is is so that they would understand the granularity. Yes, but if, as Liz is suggesting, people who have strong accents know that they do - I mean that's what I had before, and you told me to list the regions to list them. and are - Each - each one has pros and cons I mean we - we - Right. Well, I know. Right. Right. So. That's true. Yeah last week - last week I was sort of r- arguing for having it wide open, but then everybody said "Oh, no, but then it will be hard to interpret because some people will say Cincinnati and some will say Ohio". And. I mean I had it wide open last week and - and you said TIMIT. What if we put in both? Yeah. Yeah. And - Would people - That's what the "Other" is for. No, I mean what if we put in both ways of asking them? So. One is Region and the another one is "if you had to characterize yourself - your accent, what would you say?" Won't they answer the same thing? Well they might only answer only one of the questions but if Yeah that's fine. You know. They might say "Other" for Region because they don't know what category to use Actually - Right. It just - but they might have something - because it is easier to have it open ended. And we - we might learn from what they say, as to which one's a better way to ask it. W- But - I - Cuz I really don't know. This is just a small thing but @@ um It says "Variety" and then it gives things that e- have American as one of the choices. But then it says "Region", but Region actually just applies to uh, U_S, right? Right. I mean that's why I put the "Other" in. Well, we thought about it. Ah, O_K. Yeah, O_K. We just - We sort of thought, "yes, -" S- y- y- I mean - At the last meeting, my recollection was that we felt people would have uh less - that - that there are so many types and varieties of these other languages and we are not going to have that many subjects from these different language groups and that it's a Yep. huge waste of - of space. O_K. So I mean, I - I mean the way I had it last time was Region was blank, it just said Region colon. That's what I thought. Yeah. Yeah. And - and I think that that's the best way to do it, because - because of the problems we're talking about but what we said last week, was no, put in a list, so I put in a list. So should we go back to - Maybe we can make the list a little smaller. Well, certainly dropping "Northern " I think is right, because none of us know what that is. Cuz, I mean - And keeping "Other", and then maybe this North Midland, we call it "North Midwest". South Midwest, or just - Yes I - South Midwest. Does that make sense? I - I think so. Yeah. South Midwest? That would help me - Yeah. Cuz - U- unless you're from Midland, Kansas. But. Yeah. Midland - I don't know where Midland is There's a - Or Midland - Midland - Is "Midwest" one word? Is it Midland - Midland - Midland, Texas or Midland, Kansas? I forget. Y- yeah, one w- But there's a town. Oh. in - in there. I forget what it is @@ . I don't think that's what they mean. But, yeah. So. Kansas would be South Midland. Right? Yeah. Y- yeah. And - and wouldn't - Yeah. Southern Midland. @@ So, th- And Colorado, right across the border, would be I'm from Kansas, actually. North Midland. Yeah. Colora- Oh, right. And then, the - And uh - the - dropping North, so it would be Western. It's just one big shebang, where, of course, you have huge variation in dialects, but - But you do in the others, too. So. But that's true of New England too. but so do you - Yeah. So. I mean only one - Yeah. Yeah. Yeah. Well, I shouldn't say that. I have no clue. I was going to say the only one that doesn't have a huge variety is New York City. But I have no idea whether it does or not. It does seem - I mean. I - I would U- think that these categories would be more - w- would be easier for an an- analyst to put in rather than the subject himself. I think that - that was what happened with TIMIT, was that it was an analyst. O_K. Wait a minute. Where does - Yeah, I don't know how it came from. Where does - d- w- O_K. Where - Where's - where does uh New - New York west of - west of uh New York City and Pennsylvania uh and uh So. That's New England I think. Yeah. New England N- No, it's not. Oh, no. I sort of thought they were part of the - one of the Midlands. Oh no. No, no. No. Pennsylvania is not - "Other", it goes under "Other", definitely under "Other". Well, you know, Pennsylvania has a pretty strong dialect and it's totally different than - Pennsylvania - Yeah. Pennsylvania is not New England. and uh New Jersey is not New England and Maryland is not New England and none of those are the South. O_K. So. Another suggestion. Yeah. Rather than have circle fill in forms, say "Region, open paren, E_G_ Southern comma Western comma close paren colon." O_K. O_K! That's good. Yeah. Fine by me, fine by me. Sure! I like that. Yeah. Let's just - And we'll see what we get. Yeah. We're all sufficiently tired of this Be easier on the subjects. that we're agreeing with you. I think that's fine. So. No. I think - I like that. I like that. You like it? O_K. Yeah, I do. Good. Actually, maybe we do one non-English one as well. Southern, Cockney? Yeah, and - Yeah. Is that a real accent? Sure, yeah! Yeah. How do you spell it? I think that's fine. Cockney? C_O - N_ E _Y? Yeah. You could say Liverpool. Liverpuddlian. Yeah. Actually, Liverpool doesn't l- Yeah. It's - Alright. Well. Well. I mean, pure - O_K, we'll do it that way. Actually, I like that a lot. O_K. Because that get's at both of the things we were trying to do, the granularity, and the person can just self-assess and we don't have to argue about what these regions are. That's right. Yeah. O_K. And it's easy on the subjects. Yep. Now I have one suggestion on the next section. Mm-hmm. Mm-hmm. So you have native language, you have region, and then you have time spent in English speaking country. Now, I wonder if it might be useful to have another open field saying "which one parenthesis S_ paren- closed parenthesis". Cuz if they spent time in - in Britain and America - Yes. It doesn't have to be ex- all - at all exact, just in the same open field format that you have. Yep, just which one. I think that's fine. Mm-hmm. Mm-hmm. with a - with an S_ "which one sss, optional S_. O_K. Yeah. We uh - We done? Yep. Yeah, that's good. O_K. um s- e- Any - any other uh open mike topics or should we go right to the digits? Um, did you guys get my email on the multitrans? That - O_K. Isn't that wonderful! Yeah. Excellent! Thank you! I'm s- Yeah. So. So. I - I have a version also which actually displays all the channels. I ha- It's really great. But it's hideously slow. So you - this is n- Dan's patches, Dan Ellis's patches. The - what - the ones I applied, that you can actually do are Dan's, because it doesn't slow it down. M- Fantastic! Just uses a lot of memory. So when you say " slow ", does that mean to - No, the - the one that's installed is fine. It's not slow at all. I wrote another version. Which, instead of having the one pane with the one view, It has multiple panes with the views. Yeah. Mm-hmm. But the problem with it is the drawing of those waveforms is so slow that every time you do anything it just crawls. Mm-hmm. It's really bad. It's - So, it - it's the redrawing of the w- That's a consideration. Mm-hmm. oh uh-huh, w- as you move. As you play, as you move, as you scroll. Just about anything, and it - it was so slow it was not usable. And this'll be a - So that's why I didn't install it and didn't pursue it. hav- having the multiwave will be a big help cuz - in terms of like disentangling overlaps and things, Oh yeah. that'll be a big help. So. Yeah. I think that the one Dan has is usable enough. It doesn't display the others. It displays just the mixed signal. Mm-hmm. But you can listen to any of them. That's excellent. He also has version control which is another nice Yeah. e- so you - No, he suggested that, but he didn't - e- the patches that you - It's not installed. Oh, I thought it was in one of those patches. No. No. Oh O_K. Well. So is there any hope for actually displaying Alright. the wave form? Um, not if we're going to use Tcl-T_K O_K. At least not if we're going to use Snack. I mean you would have to do something ourselves. Well, or use the one that crawls. O_K. Well, I'm - I probably would be trying to use the - whatever's there. And Why don't we - we see how Dan's works and if it - it's useful to have the - Yeah. If we really need the display - I mean. I wonder - I'm just wondering if we can display things other than the wave form. So. Suppose we have a feature - a feature stream. And it's just, you know, a - a Yeah. uni-dimensional feature, varying in time. And we want to plot that, instead of the whole wave form. I mean. Yeah. That might be faster. Right? So. Yeah. We - we could do that but that would mean changing the code. I mean this isn't a program we wrote. Yeah. This is a program that we got from someone else, O_K. and we've done patches on. O_K. Mm-hmm. Yeah. Well, I'll talk to you about it and we can see but Cou- So. Yeah. i- e- I mean, y- it's definitely great to have the other one. That's - If there was some - Is there some way to have someone write patches in something faster and - and - link it in, or something? Or is that - Not easily. I mean y- yes we could do that. You could - you can write widgets in C_. Yeah. And try to do it that way but I just don't think - it - Let's try it with Dan's and if that isn't enough, Right. we can do it otherwise. I think it is, cuz when I was playing with it, the mixed signal has it all in there. And so it's really - It's not too bad to find places in the - in the stream O_K. where things are happening. And it's also - Hmm? So I - I don't think it'll be bad. also the case that - that uh this multi-wave thing is proposed to the - So. Dan proposed it to the Transcriber central people, and it's likely that uh - So. And - and they responded favorably looks as though it will be incorporated in the future version. Oh. They said that the only reason they hadn't had the multi- the parallel uh stream one before was simply that they hadn't had time to do it. And uh Yeah. so it's likely that this - this may be entered into the ch- this central @@ . They may well have not had much demand for it. And if - if - Well that's - that's - that's true, too. Yeah. This is a - Yeah. So. You mean they could - they could do it and it would be fast enough if they do it? Or - ? a useful thing for us. Oh. No. Depends on how much work they did. I just mean - I just mean that it's - that - that his - So. Oh . This one that we now have Mm-hmm. does have the status of potentially being incorporated l- likely being incorporated into the central code. O_K. Now, tha- Now, if we develop further then, y- uh, I don't - I mean it's - I think it's a nice feature to have it - I think if - if - if one of us sat down and coded it, so that it could be displayed fast enough set that way. I'm sure they would be quite willing to incorporate it. Mm-hmm. But it's not a trivial task. O_K. Mm-hmm. Yeah. I just like the idea of it being something that's, you know, tied back into the original, so that other people can benefit from it. Yeah. Yeah. However. I also understand that you can have widgets that are very useful for their purpose and that you don't need to always go that w- route. Yeah. O_K. anyway, shall we do digits? Yeah. Yeah. Let's do digits, uh, and then we'll turn off the mikes, and then I have one other thing to discuss. O_K. I actually have to leave. So. O_K. Um. I mean I had to leave at three thirty, Uh-oh. so I can - Oh. O_K. Well, you want to go first? Or. Well, I can wait for the digits but I can't stay for the discussion I c- I have to make a call. So. O_K. Well, we'll talk to you about it - Well, should we - e- should we switch off the g- Um. Do you wanna go do digits or do you wanna just skip digits? No, I can do digits if - Then - if - But I don't wanna butt in, Alright. or something. You go ahead. But if there's something on the rest of the - I'm - I'll be around just have to make call before quarter of. So. Mm-hmm. Uh So I - Or we can talk about it. Ke- Why don't you read the digits? O_K. Alright. Yeah, why don't you read the digits and then you can go. Oh, this is the new one. Yeah. Yeah, don't - Alright. The - Don't read the old one. And the time is. O_K. This is transcript two seven five one, two seven seven zero nine two five four four O_ four O_ one six zero five nine zero seven one seven nine zero eight three eight two three four zero zero seven zero three zero six four three seven seven six eight nine eight four nine nine five one O_ zero zero eight one zero six three four four two five five nine six seven zero seven eight O_ five three three Transcript uh two seven one one dash two seven three zero. seven five seven six eight eight seven zero nine seven four four four O_ O_ one nine zero zero one six seven two two nine three four six one seven four five seven zero one five s- five six four two nine five O_ six seven eight zero seven five six O_ one five nine zero four four zero five one five two two two eight one O_ two four O_ three four five seven one eight two Transcript two six five one, two six seven O_. four five zero four seven nine seven two seven one three six five eight three O_ eight nine six zero five six O_ zero one two zero two four three eight two five two zero seven seven eight three six seven six seven nine seven one seven eight nine O_ O_ two two one three seven two five five three five seven one five three eight four eight one one zero Transcript number two six seven one dash two six nine zero five six seven O_ eight three O_ nine two two O_ three six O_ seven O_ four zero six one two seven one O_ one two four three four zero six seven two six four seven eight seven four eight three O_ nine O_ zero one zero eight nine nine three one nine four two seven seven five five seven five two Transcript two six nine one, two seven one O_. six nine seven eight nine zero zero two six eight two two four one four zero eight two three three three three four nine two four five zero six five six O_ six eight four one three four nine three four O_ eight six zero one two three five two seven one eight four eight six six six two O_ seven one O_K O_K. Turn it off. But wait till he - O_K. And