<html xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=Windows-1252">
<meta name="Generator" content="Microsoft Word 15 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
{font-family:"Cambria Math";
panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
{font-family:Aptos;
panose-1:2 11 0 4 2 2 2 2 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0cm;
font-size:10.0pt;
font-family:"Aptos",sans-serif;}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:blue;
text-decoration:underline;}
span.EmailStyle19
{mso-style-type:personal-reply;
font-family:"Aptos",sans-serif;
color:windowtext;}
.MsoChpDefault
{mso-style-type:export-only;
font-size:10.0pt;
mso-ligatures:none;}
@page WordSection1
{size:612.0pt 792.0pt;
margin:72.0pt 72.0pt 72.0pt 72.0pt;}
div.WordSection1
{page:WordSection1;}
--></style>
</head>
<body lang="EN-GB" link="blue" vlink="purple" style="word-wrap:break-word">
<div class="WordSection1">
<p class="MsoNormal"><span style="font-size:11.0pt">Hi, <o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt">chiming in as well � I tried this recently with one of my papers and got the impression that it has the obvious LLM problems of hallucinating, which can be particularly problematic when one uses the podcast
to get a summary of articles (our recent paper with Stuart and Marina on <a href="https://doi.org/10.1145/3610977.3634936">
Autonomous Robots on Public Streets</a> was glossed as being about the trolley problem, which is quite a stretch).
<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt">I share the impression about the �who knows what� problem. Both voices contribute snippets of information, but they don�t seem to necessarily build on each other�s contribution, there is no mutual understanding,
even if it sounds like that on the surface (even Clark�s grounding model would apply here, which you probably are familiar with, Roger?).
<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt">It also strikes me how the prosody of these AI voices becomes so perfect that it gets boring, similar to reading AI generated texts, it sounds amazing for a while, but after some minutes I find it difficult
to remember what I just heard. <o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt">Best wishes,<br>
Hannah<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt"><o:p> </o:p></span></p>
<div id="mail-editor-reference-message-container">
<div>
<div>
<div style="border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0cm 0cm 0cm">
<p class="MsoNormal" style="margin-bottom:12.0pt"><b><span style="font-size:12.0pt;color:black">From:
</span></b><span style="font-size:12.0pt;color:black">emcai <emcai-bounces@conversationanalysis.org> on behalf of Stuart Reeves via emcai <emcai@conversationanalysis.org><br>
<b>Date: </b>Friday, 4 October 2024 at 18:38<br>
<b>To: </b>Magnus Hamann <M.Hamann@lboro.ac.uk><br>
<b>Cc: </b>L.E.M. de Rijk, (Lynn) via emcai <emcai@conversationanalysis.org><br>
<b>Subject: </b>Re: [emcai] Podcasts generated using Google's NotebookLM<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:11.0pt">I guess K+/K- at least has a clear computational mapping :)<br>
<br>
On Formal Structures of Practical Actions has a pretty hilarious summary:<br>
<a href="https://notebooklm.google.com/notebook/9c1218c4-3d43-430a-a5af-13aa831de964/audio">https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fnotebooklm.google.com%2Fnotebook%2F9c1218c4-3d43-430a-a5af-13aa831de964%2Faudio&data=05%7C02%7Channah.pelikan%40liu.se%7C5c87fdedbc6546a942be08dce492f70d%7C913f18ec7f264c5fa816784fe9a58edd%7C0%7C0%7C638636567083479922%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=YEO16gi73WNumSLfmFR%2FXTwYwSnUPA1Rbo%2BwJcVnFkQ%3D&reserved=0</a><br>
<br>
What's interesting is that what is generated is almost 'keyword' list based clearly in terms of what has the strongest signal in the model e.g., specifically:<br>
<br>
- Roses's gloss<br>
- Indexical words<br>
- Formulations<br>
- Context-free descriptions<br>
<br>
<br>
> On 3 Oct 2024, at 13:32, Magnus Hamann via emcai <emcai@conversationanalysis.org> wrote:<br>
><br>
> Hi Hendrik,<br>
><br>
> This was also a thing I noticed. That the AI keeps interchanging their roles of K+ and K- (as some would maybe dub it).<br>
><br>
> -----Original Message-----<br>
> From: emcai <emcai-bounces@conversationanalysis.org> On Behalf Of Hendrik Buschmeier via emcai<br>
> Sent: 03 October 2024 13:16<br>
> To: L.E.M. de Rijk, (Lynn) via emcai <emcai@conversationanalysis.org><br>
> Subject: Re: [emcai] Podcasts generated using Google's NotebookLM<br>
><br>
> ** THIS MESSAGE ORIGINATED OUTSIDE LOUGHBOROUGH UNIVERSITY **<br>
><br>
> ** Be wary of links or attachments, especially if the email is unsolicited or you don't recognise the sender's email address. **<br>
><br>
> Hi Roger,<br>
><br>
> No real analysis, just an impression � seconding also Magnus analysis:<br>
><br>
> The �hosts� communicative feedback is off and makes it unclear who knows what. In their listener behavior they produce newsmakers, as if what the other says is unknown to them, but then, in the next turn, they just take over and seemingly have all the information,
and suddenly the other seems clueless. I think the model is missing that the two speakers are independent entities with asymmetric knowledge. To me this shows how important representations of the self and the other are, even when just producing something
that �just generates� output and is not interactive.<br>
><br>
> More generally I question the merit of such applications. There are more than enough podcasts. Some people mentioned that the thing is prone to halucination as well.<br>
><br>
> From a scientific network perspective, we could think about an in depth analysis of one of these "dialogues" looking at various conversational phenomena that are produced.<br>
><br>
> Cheers<br>
><br>
> Hendrik<br>
><br>
><br>
><br>
>> On 3. Oct 2024, at 11:46, Nicolas Rollet via emcai <emcai@conversationanalysis.org> wrote:<br>
>><br>
>> Hi,<br>
>><br>
>> pauses are over-marked too, as if it's based on the fact that pauses is the key-trigger for continuers, with remind me of a very basic definition of turn in computer science, no ?<br>
>><br>
>> Best wishes<br>
>><br>
>> Nicolas Rollet<br>
>> T�l�com Paris<br>
>> Department of Economics and Social Sciences INTERACT Institut<br>
>> polytechnique de Paris<br>
>> I3 UMR 9217 � CNRS<br>
>> Site i3<br>
>> Encyclop�die de la parole<br>
>> (+33)6 88 75 86 28<br>
>> bureau 3A259<br>
>><br>
>> De: "L.E.M. de Rijk, (Lynn) via emcai"<br>
>> <emcai@conversationanalysis.org><br>
>> �: "L.E.M. de Rijk, (Lynn) via emcai" <emcai@conversationanalysis.org><br>
>> Envoy�: Mercredi 2 Octobre 2024 18:59:11<br>
>> Objet: Re: [emcai] Podcasts generated using Google's NotebookLM<br>
>><br>
>> I also immediately noticed the massive overuse of continuers,<br>
>> acknowledgments and other response tokens. But I will say, I just<br>
>> tried this thing with one of our papers, and it worked very smoothly<br>
>> there, generating quite a realistic experience. The "podcast" really<br>
>> captures the majority of the contents of the paper accurately and even<br>
>> offers comprehensible summary descriptions of what happens in the data<br>
>> fragments :-O<br>
>><br>
>> Quite fascinating, I must say! Thanks for sharing this, Roger.<br>
>> Best,<br>
>> Uwe<br>
>><br>
>><br>
>> Am 02.10.2024 um 18:51 schrieb Magnus Hamann via emcai:<br>
>> Just a first noticing:<br>
>> There is definitely something with the sequential distribution of "continuers" that is a little off.<br>
>> That said, this is definitely an interesting tool (thinking about students with dyslexia here).<br>
>><br>
>><br>
>><br>
>> Kind regards,<br>
>> Magnus Hamann<br>
>><br>
>> Informal office hour every Monday, Tuesday (both online), or Thursday<br>
>> (in my office) 11-12<Outlook-ypjhe4zs.png><br>
>><br>
>> Book time to meet with me<br>
>> <Outlook-ukk32cwg.png>From: emcai<br>
>> <emcai-bounces@conversationanalysis.org> on behalf of Roger K Moore<br>
>> via emcai <emcai@conversationanalysis.org><br>
>> Sent: 02 October 2024 16:54<br>
>> To: emcai@conversationanalysis.org <emcai@conversationanalysis.org><br>
>> Subject: [emcai] Podcasts generated using Google's NotebookLM<br>
>> ** THIS MESSAGE ORIGINATED OUTSIDE LOUGHBOROUGH UNIVERSITY **<br>
>> ** Be wary of links or attachments, especially if the email is<br>
>> unsolicited or you don't recognise the sender's email address. ** Hi<br>
>> All,<br>
>><br>
>> In case you are not aware, the latest AI development from Google provides the ability to generate a podcast-style conversation between two 'individuals' automatically from an uploaded document (e.g. a scientific paper)? The results are astounding, both
in terms of the speech and the dialogue.<br>
>><br>
>> Here are two examples I generated based on ...<br>
>> �<br>
>> my recent VIHAR-24 paper on "What Needs to be Known in Order to Perform a Meaningful Scientific Comparison Between Animal Communications and Human Spoken Language", and<br>
>> � the famous "Chicken Chicken Chicken" paper by Doug Zongker.<br>
>> I wonder how these AI-generated interactions stack-up from a CA perspective?<br>
>><br>
>> Best wishes<br>
>> Roger<br>
>><br>
>> ----------------------------------------------------------------------<br>
>> --------------------------- Prof ROGER K MOORE* BA(Hons) MSc PhD FIOA<br>
>> FISCA MIET SMIEEE<br>
>><br>
>> Deputy Head of School<br>
>> Chair of Spoken Language Processing<br>
>> Head of Speech & Hearing Research Group (SpandH) Vocal Interactivity<br>
>> Lab (VILab), Sheffield Robotics School of Computer Science, UNIVERSITY<br>
>> OF SHEFFIELD Regent Court, 211 Portobello, Sheffield, S1 4DP, UK<br>
>><br>
>> * Winner of the 2016 Antonio Zampolli Prize for "Outstanding<br>
>> Contributions to the Advancement of Language Resources & Language<br>
>> Technology Evaluation within Human Language Technologies"<br>
>><br>
>> e-mail: r.k.moore@sheffield.ac.uk<br>
>> web: <a href="http://staffwww.dcs.shef.ac.uk/people/R.K.Moore/">https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fstaffwww.dcs.shef.ac.uk%2Fpeople%2FR.K.Moore%2F&data=05%7C02%7Channah.pelikan%40liu.se%7C5c87fdedbc6546a942be08dce492f70d%7C913f18ec7f264c5fa816784fe9a58edd%7C0%7C0%7C638636567083500630%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=aDHBwvmKc71%2Bhkn6I3ITm2XxczT5LrDSjVQousef75c%3D&reserved=0</a><br>
>> twitter: @rogerkmoore<br>
>> Tel: +44 (0) 11422 21807<br>
>> Fax: +44 (0) 11422 21810<br>
>> Mob: +44 (0) 7910 073631<br>
>><br>
>> Editor-in-Chief: COMPUTER SPEECH AND LANGUAGE<br>
>> (<a href="http://www.journals.elsevier.com/computer-speech-and-language/">https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.journals.elsevier.com%2Fcomputer-speech-and-language%2F&data=05%7C02%7Channah.pelikan%40liu.se%7C5c87fdedbc6546a942be08dce492f70d%7C913f18ec7f264c5fa816784fe9a58edd%7C0%7C0%7C638636567083511144%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=3j4WWv%2FHQlUgRJQj2nPg%2FtYGD2GNVXxww3KedW4GY2A%3D&reserved=0</a>)<br>
>> ----------------------------------------------------------------------<br>
>> ----------------------<br>
>><br>
>><br>
>> --<br>
>> emcai mailing list<br>
>> emcai@conversationanalysis.org<br>
>> <a href="http://conversationanalysis.org/mailman/listinfo/emcai_conversationana">
https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fconversationanalysis.org%2Fmailman%2Flistinfo%2Femcai_conversationana&data=05%7C02%7Channah.pelikan%40liu.se%7C5c87fdedbc6546a942be08dce492f70d%7C913f18ec7f264c5fa816784fe9a58edd%7C0%7C0%7C638636567083521555%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=v185%2BYzxRzt2dcbRfMPzk40OfeAE5ZrwANQKqICIInQ%3D&reserved=0</a><br>
>> lysis.org<br>
>> --<br>
>> emcai mailing list<br>
>> emcai@conversationanalysis.org<br>
>> <a href="http://conversationanalysis.org/mailman/listinfo/emcai_conversationana">
https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fconversationanalysis.org%2Fmailman%2Flistinfo%2Femcai_conversationana&data=05%7C02%7Channah.pelikan%40liu.se%7C5c87fdedbc6546a942be08dce492f70d%7C913f18ec7f264c5fa816784fe9a58edd%7C0%7C0%7C638636567083532158%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=nw7bmXaWFl4hPYHcYB5ziGL4cT%2FsA6cJMHeqh2vX4sI%3D&reserved=0</a><br>
>> lysis.org<br>
><br>
><br>
> --<br>
> Hendrik Buschmeier<br>
> Digital Linguistics Lab<br>
> Faculty of Linguistics and Literary Studies, Bielefeld University <a href="https://purl.org/net/hbuschme">
https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpurl.org%2Fnet%2Fhbuschme&data=05%7C02%7Channah.pelikan%40liu.se%7C5c87fdedbc6546a942be08dce492f70d%7C913f18ec7f264c5fa816784fe9a58edd%7C0%7C0%7C638636567083542144%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=t5n4qLa9BBQ4f6fBydSIPhgrxSUxZp34fbnBei3i4tc%3D&reserved=0</a><br>
><br>
><br>
><br>
> --<br>
> emcai mailing list<br>
> emcai@conversationanalysis.org<br>
> <a href="http://conversationanalysis.org/mailman/listinfo/emcai_conversationanalysis.org">
https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fconversationanalysis.org%2Fmailman%2Flistinfo%2Femcai_conversationanalysis.org&data=05%7C02%7Channah.pelikan%40liu.se%7C5c87fdedbc6546a942be08dce492f70d%7C913f18ec7f264c5fa816784fe9a58edd%7C0%7C0%7C638636567083552137%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=a%2FYbWtK5tO481vTsz5F4R7oiABM0yX8Gk9Zc%2FBmiFEc%3D&reserved=0</a><br>
> --<br>
> emcai mailing list<br>
> emcai@conversationanalysis.org<br>
> <a href="http://conversationanalysis.org/mailman/listinfo/emcai_conversationanalysis.org">
https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fconversationanalysis.org%2Fmailman%2Flistinfo%2Femcai_conversationanalysis.org&data=05%7C02%7Channah.pelikan%40liu.se%7C5c87fdedbc6546a942be08dce492f70d%7C913f18ec7f264c5fa816784fe9a58edd%7C0%7C0%7C638636567083562081%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=8UDbx9DRuFfv%2FpZJzJhi04YI0m0FRulcD3ZLLLopguk%3D&reserved=0</a><br>
<br>
Stuart Reeves | Mixed Reality Lab, School of Computer Science (C15), University of Nottingham<br>
<a href="http://www.cs.nott.ac.uk/~str">https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.cs.nott.ac.uk%2F~str&data=05%7C02%7Channah.pelikan%40liu.se%7C5c87fdedbc6546a942be08dce492f70d%7C913f18ec7f264c5fa816784fe9a58edd%7C0%7C0%7C638636567083571999%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=sGNGzCNHy2eSTN7nAD1%2BsEF5MfPrID6vMCS1yJRQb4c%3D&reserved=0</a><br>
<br>
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please contact the sender and delete the email and attachment. Any views or opinions expressed by the
author of this email do not necessarily reflect the views of the University of Nottingham. Email communications with the University of Nottingham may be monitored where permitted by law.<br>
-- <br>
emcai mailing list<br>
emcai@conversationanalysis.org<br>
<a href="http://conversationanalysis.org/mailman/listinfo/emcai_conversationanalysis.org">https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fconversationanalysis.org%2Fmailman%2Flistinfo%2Femcai_conversationanalysis.org&data=05%7C02%7Channah.pelikan%40liu.se%7C5c87fdedbc6546a942be08dce492f70d%7C913f18ec7f264c5fa816784fe9a58edd%7C0%7C0%7C638636567083582721%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=BmXnG6alk3En9gbf8e67Lyw1aKzjr%2Bn%2BOKNlLeQ9hP4%3D&reserved=0</a><o:p></o:p></span></p>
</div>
</div>
</div>
</div>
</div>
</body>
</html>