From dana@emu.com Mon May 17 09:04:08 1993 Received: from uu6.psi.com by marsh.cs.curtin.edu.au with SMTP id AA06253 (5.65c/IDA-1.4.4 for ); Tue, 18 May 1993 07:23:13 +0800 Received: from virgil.UUCP by uu6.psi.com (5.65b/4.0.071791-PSI/PSINet) via UUCP; id AA22096 for ; Mon, 17 May 93 19:12:59 -0400 Received: from chronos.emu by virgil (4.1/SMI-4.1) id AA12132; Mon, 17 May 93 16:04:11 PDT Received: by chronos.emu (4.1/SMI-4.1) id AA02989; Mon, 17 May 93 16:04:08 PDT Date: Mon, 17 May 93 16:04:08 PDT From: dana@emu.com (Dana Massie) Message-Id: <9305172304.AA02989@chronos.emu> To: val@cs.curtin.edu.au Subject: interest in paper... Status: RO Hello - I saw your listing in COMP.DSP about facial animation, and i have a friend who is very interested in learning about this technology. My friend is Rand Wetherwax, and Rand has been doing some work with facial animaation using real time inputs to control "virtual puppets" which are 3-d face models. He is involved in other projects that have involved synching dialog and graphics, and I know that he is curious about your work from my description. He does not have a decent email connection, though. If you can possibly send him a paper copy of some of your work, I know would like to see it (i talked to him about it), and feel free to call him if you want to share ideas. His address is Rand Wetherwax 30037 Rainbow Dr. Val Verde, CA 91350 USA tel: 213-465-4855 thanks for you time - - dana dana c. massie E-mu Systems, Inc. P.O. Box 660015 Scotts Valley, CA 95067-0015 408.438.1921 dana@emu.com From bilbo@tdisys.uucp Mon May 17 04:54:23 1993 Received: from corton.inria.fr by marsh.cs.curtin.edu.au with SMTP id AA19358 (5.65c/IDA-1.4.4 for ); Mon, 17 May 1993 18:34:56 +0800 Received: from tdisys.UUCP by corton.inria.fr (5.65c8d/92.02.29) via Fnet-EUnet id AA02173; Mon, 17 May 1993 11:35:00 +0200 (MET) Received: by tdisys.tdisys.fr, Mon, 17 May 93 10:54:23 MDT Date: Mon, 17 May 93 10:54:23 MDT From: bilbo@tdisys.uucp (Jean-Luc Martinez) Message-Id: <9305170854.AA08147@tdisys.tdisys.fr> To: val@cs.curtin.edu.au Subject: Speech recognition and facial animation Status: RO Hi, I'm working at TDI on polygonal modeling and I've currently a special interest in 3D morphing, so I'll read your report (3D morphing or Image morphing ?) and send you my remarks if any. We are also interesting in character and facial animation and our animation team will read your thesis. We are in contact with a girl who have done her thesis in facial animation and currently work at Philadelphia on speech phonemes recognition. > Marie-Luce VIAUD, E-mail: luce@bora.inria.fr Thanks for your informations, Jean-Luc MARTINEZ R&D - Broadcast Team Thomson Digital Image 75018 PARIS - FRANCE Phone: (33-1) 43 87 58 58 Fax: (33-1) 43 87 61 11 E-mail: bilbo@tdisys.uucp bilbo@thomson-di.fr From geigel@seasva.gwu.edu Mon May 17 06:43:14 1993 Received: from habanero.seas.gwu.edu by marsh.cs.curtin.edu.au with SMTP id AA24609 (5.65c/IDA-1.4.4 for ); Mon, 17 May 1993 22:43:28 +0800 Received: from cayenne (cayenne.seas.gwu.edu) by habanero.seas.gwu.edu (4.1+rlw3va) id AA00420; Mon, 17 May 93 10:43:16 EDT From: Joseph M. Geigel Received: by cayenne id AA00422; Mon, 17 May 93 10:43:14 EDT Date: Mon, 17 May 93 10:43:14 EDT Message-Id: <9305171443.AA00422@cayenne> To: val@cs.curtin.edu.au Subject: Re: Speech recognition and facial animation Newsgroups: comp.graphics In-Reply-To: Organization: George Washington University Status: RO Valerie... Thanks for making all of your facial stuff available. > >I'm doing my PhD in speech driven facial animation and would like to >hear from anyone else working in the area. My main interest is in the >speech recognition area, realtime recognition of speech units (eg. >phonemes) rather than word comprehension. There have been some work done, as you probably know, on mapping phonemes to facial parameters (Magnenat-Thalman, Thallman for one), but all that I have seen assumes that somehow you have already have the phonemes. I hate to send a "me too" reply...but if you do find any info, could you possibly forward or post it? I, too am interested in this area. (although not to the extent that you probably are. My interests lie in sound/animation synchonization, with lip syncing being a part of that). Thanks again, -- jogle geigel@seasva.gwu.edu From welsh_w_j@bt-web.bt.co.uk Tue May 11 10:44:34 1993 Received: from zaphod.axion.bt.co.uk by marsh.cs.curtin.edu.au with SMTP id AA06148 (5.65c/IDA-1.4.4 for ); Tue, 11 May 1993 16:46:07 +0800 Message-Id: <199305110846.AA06148@marsh.cs.curtin.edu.au> Received: from bt-web.bt.co.uk by zaphod.axion.bt.co.uk via DECnet inbound channel id ; Tue, 11 May 1993 09:44:34 +0100 X-Vms-To: R11F::CS.CURTIN.EDU.AU::VAL To: val@cs.curtin.edu.au From: welsh_w_j@bt-web.bt.co.uk Subject: greetings Date: Tue, 11 May 1993 09:44:34 +0100 Status: RO Hello Val, How is your project coming along? Gareth Lee contacted BT on your behalf to ask whether we had any papers which might be useful to you. I remember sending you some stuff before but I don't remember exactly what it was. Did I send you anything on the speech recognition algorithm? Anyway, it must have been a long time ago since I was in communication with you last. I spent five months of last year working at the MIT Media lab doing work on image analysis as part of a research contract we have with them, I had a pretty good time. Well, keep in touch. Bill Welsh From thinman@netcom.com Sat May 15 11:48:46 1993 Received: from netcom4.netcom.com by marsh.cs.curtin.edu.au with SMTP id AA18193 (5.65c/IDA-1.4.4 for ); Sun, 16 May 1993 09:48:34 +0800 Received: by netcom4.netcom.com (5.65/SMI-4.1/Netcom) id AA20853; Sat, 15 May 93 18:48:46 -0700 Date: Sat, 15 May 93 18:48:46 -0700 From: thinman@netcom.com (Technically Sweet) Message-Id: <9305160148.AA20853@netcom4.netcom.com> To: val@cs.curtin.edu.au Subject: Re: Speech recognition for facial animation Newsgroups: comp.dsp References: Status: RO More face stuff: the "rapid face construction lab" is part of a Sony Research project on mobile networked computers, personal assistants, and the OS needed to run it all. The papers are in: scslwide.sony.co.jp:CSL/CSL-Papers -- Lance Norskog thinman@netcom.com Data is not information is not knowledge is not wisdom. From zilla@ccrl.nj.nec.com Fri May 14 09:01:20 1993 Received: from overdrive.ccrl.nj.nec.com by marsh.cs.curtin.edu.au with SMTP id AA20928 (5.65c/IDA-1.4.4 for ); Sat, 15 May 1993 01:02:15 +0800 Received: by overdrive.ccrl.nj.nec.com (4.1/EMS1.7-930224.09) id AA21287(overdrive.ccrl.nj.nec.com); Fri, 14 May 93 13:01:32 EDT From: zilla@ccrl.nj.nec.com (John Lewis) Received: by trillian (4.1/CNC-Client) id AA04760; Fri, 14 May 93 13:01:20 EDT Date: Fri, 14 May 93 13:01:20 EDT Message-Id: <9305141701.AA04760@trillian> To: val@cs.curtin.edu.au Subject: speech-driven face Status: RO Hi, I've been working on speech-driven faces occasionally. You probably already have these references. I might be able to provide an electronic version of the 91 article if your interested. Let's keep in touch, we might be able to share some useful info. j.p.lewis/nec research ----------------------- @InProceedings{Lewi87, author = "Lewis, J.P. and Parke, F.I.", title = "Automated lipsynch and speech synthesis for character animation", booktitle = "Proceedings Human Factors in Computing Systems and Graphics Interface '87", year = "1987", OPTeditor = "", pages = "143-147", OPTorganization = "", OPTpublisher = "", OPTaddress = "", OPTmonth = "", OPTnote = "" } @Article{Lewi91, author = "Lewis, John", title = "Automated Lip-sync: Background and Techniques", journal = "The Journal of Visualization and Computer Animation", year = "1991", volume = "2", number = "4", pages = "118--122", month = "October--December", OPTnote = "" } From rab@cs.cmu.edu Fri May 14 19:00:21 1993 Received: from CS.CMU.EDU by marsh.cs.curtin.edu.au with SMTP id AA00588 (5.65c/IDA-1.4.4 for ); Sat, 15 May 1993 11:02:15 +0800 Received: from RAB-RTR.SPEECH.CS.CMU.EDU by CS.CMU.EDU id aa16724; 14 May 93 23:01:46 EDT Received: by rab.speech.cs.cmu.edu (NX5.67c/NX3.0S) id AA03342; Fri, 14 May 93 23:00:21 -0400 Date: Fri, 14 May 93 23:00:21 -0400 From: Robert Brennan Message-Id: <9305150300.AA03342@rab.speech.cs.cmu.edu> Received: by NeXT.Mailer (1.87.1) Received: by NeXT Mailer (1.87.1) To: val@cs.curtin.edu.au (Valerie Hall) Subject: Re: Speech recognition for facial animation Status: RO Yes, work has been done in this area. Cambridge Animation Systems has a very nice animation product that incorporates speech recognition to aid in facial animation. The speech recognition is based on work I did for them. Contact Peter Florence (pete@cam-ani.co.uk) for more details. If you have questions related to the recognition component, I could answer those for you provided that I do not disclose details of the specific implementation that I did for Cambridge Animation Systems. Bob Brennan In comp.dsp article you wrote: > > > Hi, > > > I'm doing my PhD in speech driven facial animation and would like to > hear from anyone else working in the area. My main interest is in the > speech recognition area, realtime recognition of speech units (eg. > phonemes) rather than word comprehension. > > If anyone knows of any useful references, leads or keywords for > searches, I'd be glad to hear from you. > > For anyone interested in facial animation, I have some information in > my ftp area. There is also some info on morphing, for those who > haven't found it yet. > > > Thanks in advance, > > > Valerie Hall > Doctoral Student > Curtin University of Technology > Perth, Western Australia > > ------------------------README file----------------------------- > > 1) my thesis on Facial Animation (helpful for anyone > looking for references in the area) > > 2) A model of half a face, data points and topology. > > 3) my report on Morphing (an introduction to morphing) > > 4) responses to my questions about morphing. > > 5) the results of a survey I did on user interfaces. > > 6) the survey responses. > > 7) a technical report by Steve Franks at SONY on facial > animation. > > You can do whatever you like with them. Let me know what you think, > constructive criticism will be greatly appreciated. > > > site: marsh.cs.curtin.edu.au > > dir: pub/graphics/bibliography/Facial_Animation > files: > > **** new files **** > > thesis.ps.Z (facial animation thesis, Nov '92) > thesis.txt.Z (text version of thesis, Nov '92) > face2face.ps.Z (paper submitted to local conference) > errors (error report for thesis and other files) > > **** older files **** > > proposal.ps.Z (My thesis proposal - 1.5M) > proposal.txt.Z (My thesis proposal - text file) > ann_biblio.ps.Z (annotated bibliography) > ann_biblio.txt.Z (annotated biblio - text file) > valerie.info (info on data for half a face) > valerie.st1 (3-D data for half a face) > valerie.top (topology for facial data) > SCSL-TR-92-010.ps.Z (Steve Franks' facial animation paper) > > dir: pub/graphics/bibliography/Morph > files: > **** new files **** > > morph_prog (info on a morphing prog for windows) > morphine.sites (sites where morphine can be found) > warping.ps.Z (paper on image warping by > > **** older files **** > > morph_intro.ps.Z (My morphing write-up - 1.5M) > morph_intro.txt.Z (My morphing write-up - text) > m_responses.Z (Responses to morphing questions) > > dir: pub/graphics/bibliography/UI > files: survey.ps.Z (Writeup for user interfaces survey) > survey.ps.Z (Textfile writeup for UI survey) > responses.Z (Responses to UI questionaire) > > > Enjoy! > > -- > ----------------val@marsh.cs.curtin.edu.au-------------- /\_/\ > You can't know what love isn't... FOXY / O O \ > ...until you know what love is. LADY \ \./ / > ----------------------------------------My other car is a broom! From lon@cwi.nl Fri May 14 10:10:47 1993 Received: from charon.cwi.nl by marsh.cs.curtin.edu.au with SMTP id AA10609 (5.65c/IDA-1.4.4 for ); Fri, 14 May 1993 18:11:03 +0800 Received: from guppie.cwi.nl by charon.cwi.nl with SMTP id AA16938 (5.65b/3.8/CWI-Amsterdam); Fri, 14 May 1993 12:10:48 +0200 Received: by guppie.cwi.nl id AA08650 (5.65b/3.8/CWI-Amsterdam); Fri, 14 May 1993 10:10:47 GMT Date: Fri, 14 May 1993 10:10:47 GMT From: Lon.Barfield@cwi.nl Message-Id: <9305141010.AA08650=lon@guppie.cwi.nl> To: val@cs.curtin.edu.au Subject: Re: Speech recognition and facial animation Newsgroups: comp.graphics References: Status: RO In comp.graphics you write: >I'm doing my PhD in speech driven facial animation and would like to >hear from anyone else working in the area. My main interest is in the >speech recognition area, realtime recognition of speech units (eg. >phonemes) rather than word comprehension. There is someone here doing similar stuff, bulding up basic mouth shapes for basic phonemes and then using this info to drive computer animations. Ask her about the talking pear! patsy@cwi.nl cheers -- Lon -- Lon Barfield, CWI, Amsterdam, The Netherlands lon@cwi.nl +31 20 592 4140 From jack@cwi.nl Mon May 17 16:36:16 1993 Received: from charon.cwi.nl by marsh.cs.curtin.edu.au with SMTP id AA21819 (5.65c/IDA-1.4.4 for ); Mon, 17 May 1993 20:36:24 +0800 Received: from schelvis.cwi.nl by charon.cwi.nl with SMTP id AA06091 (5.65b/3.8/CWI-Amsterdam); Mon, 17 May 1993 14:36:16 +0200 Received: by schelvis.cwi.nl id AA06766 (5.65b/3.8/CWI-Amsterdam); Mon, 17 May 1993 14:36:16 +0200 Date: Mon, 17 May 1993 14:36:16 +0200 From: Jack.Jansen@cwi.nl Message-Id: <9305171236.AA06766=jack@schelvis.cwi.nl> To: val@cs.curtin.edu.au Subject: Re: Speech recognition and facial animation Newsgroups: comp.sys.sgi.misc References: Status: RO Valerie, Patricia Griffin here at CWI is doing work quite similar to yours. She is looking at mouth shapes corresponding to phonemes, and uses that to make various objects talk. (Well, that is my understanding of it, don't be surprised if it is grossly incorrect:-). You can contact her as patsy@cwi.nl. She is out of the country at the moment, though, so it might take a while before she replies. Success, -- -- Jack Jansen | If I can't dance I don't want to be part of Jack.Jansen@cwi.nl | your revolution -- Emma Goldman uunet!cwi.nl!jack G=Jack;S=Jansen;O=cwi;PRMD=surf;ADMD=400net;C=nl From mmt@ccu1.aukuni.ac.nz Sun May 19 04:05:05 1993 Received: from ccvcom.aukuni.ac.nz by marsh.cs.curtin.edu.au with SMTP id AA13277 (5.65c/IDA-1.4.4 for ); Tue, 18 May 1993 12:08:20 +0800 Received: from ccu1.aukuni.ac.nz by aukuni.ac.nz (PMDF #2864 ) id <01GYBRPLG0MO95NFVX@aukuni.ac.nz>; Tue, 18 May 1993 16:07:05 +1200 Received: by ccu1.aukuni.ac.nz with 920330.SGI/920826.1 id ; Tue, 18 May 93 16:05:05 +1200 Date: 18 May 1993 16:05:05 +1200 From: mmt@ccu1.aukuni.ac.nz (Mark Thomson) Subject: Re: Speech recognition for facial animation To: val@cs.curtin.edu.au Message-Id: <9305180405.AA04070@ccu1.aukuni.ac.nz> Content-Transfer-Encoding: 7BIT Newsgroups: comp.dsp References: Status: RO In comp.dsp you write: >I'm doing my PhD in speech driven facial animation and would like to >hear from anyone else working in the area. My main interest is in the >speech recognition area, realtime recognition of speech units (eg. >phonemes) rather than word comprehension. If I understand correctly what you're doing then it sounds like it's related to articulatory synthesis, i.e. the problem of deriving appropriate parameters to control a model which represents the physical characteristics of the vocal system. These parameters can be the cross-sectional areas of the vocal tract at different points along its length, or these may be equivalently represented by the positions of the articulators such as the jay and tongue etc. Two papers which describe the relationship between articulator positions and cross-sectional areas are : C.H. Coker, A Model of Articulatory Dynamics and Control, Proc. IEEE, April 1976, pp452-460. P. Mermelstein, Articulatory Model for the Study of Speech Production, J. Acoust. Soc. Am., vol 53, No. 4, pp1070-1082, 1973. A paper which describes a model of speech production in the vocal system is M.M. Sondhi and J. Schroeter, "A Hybrid Time Frequency Domain Articulatory Speech Synthesiser" IEEE Trans Acoust. Speech, Signal Processing, vol ASSP-35, No. 7, pp 955-967, 1987 There have been a number of papers which deal with the problem of controlling the model described in Sondhi and Schroeter's paper. The most recent (to my knowledge) is M.G. Rahim et al., "Acoustic to Articulatory Parameter Mapping Using an Assembly of Neural Networks", Proc. IEEE Int. Conf on Acoustics, Speech and Signal Processing, 1991, pp485-488. Hope this helps, Mark Thomson Electrical and Electronic Engineering University of Auckland From dpr@ohm.york.ac.uk Sun May 16 14:55:52 1993 Received: from glenlivet.ohm.york.ac.uk by marsh.cs.curtin.edu.au with SMTP id AA28450 (5.65c/IDA-1.4.4 for ); Sun, 16 May 1993 20:55:59 +0800 Received: by glenlivet.ohm.york.ac.uk (/\==/\ Smail3.1.28.1 #28.41) id m0nuiFs-000F0GC; Sun, 16 May 93 13:55:52 +0100 (BST) Message-Id: Date: Sun, 16 May 93 13:55:52 +0100 (BST) From: "David P Rossiter" To: val@cs.curtin.edu.au Subject: Re: Speech recognition and facial animation Newsgroups: comp.sys.sgi.misc References: Status: RO Hi Well, my work isn't exactly the same as yours, but yours is a subset of mine (if you see what I mean). I am the Research Associate on 3 year project 'Real time visual displays for professional voice development'. I'm half way through now. Using an SGI as a base, we're developing graphics programs that take the audio (or larynx) signal and present the information visually in an appropriate form for the developing voice user. Hopefully, anyway. There is a paper entitled 'real-time visual displays for singing development' I wrote recently which I could ftp somewhere if you were interested (postscript). That describes 3 of our displays and will give you an idea of what we're about. I haven't had a look at your ftp stuff yet, but generally speaking we aren't looking at your specific area, although obviously your research is of great interest. Are you SGI based? Any papers yet? etc, etc - I'll have a look at your stuff before asking anything more, but generally I'd love to be kept in touch with any developments you come up with. Hope things go well Dave From jcr@its.com Sat May 15 12:28:18 1993 Received: from uu2.psi.com by marsh.cs.curtin.edu.au with SMTP id AA16856 (5.65c/IDA-1.4.4 for ); Sun, 16 May 1993 07:35:42 +0800 Received: by uu2.psi.com (5.65b/4.0.071791-PSI/PSINet) via UUCP; id AA00581 for ; Sat, 15 May 93 19:11:22 -0400 From: jcr@its.com (John C. Randolph) Received: from endor (endor.ARPA) by its.com (NeXT-1.0 (From Sendmail 5.52)/3.2.083191-(Information Technology Solutions)) id AA02788; Sat, 15 May 93 17:28:05 CDT Message-Id: <9305152228.AA02788@its.com> Received: by endor (NX5.67d/NX3.0X) id AA00542; Sat, 15 May 93 17:28:18 -0500 Date: Sat, 15 May 93 17:28:18 -0500 Received: by NeXT.Mailer (1.94) Received: by NeXT Mailer (1.94) To: val@cs.curtin.edu.au (Valerie Hall) Subject: Re: Speech recognition and facial animation Reply-To: jcr@its.com Status: RO You wrote: > I'm doing my PhD in speech driven facial animation and would like to > hear from anyone else working in the area. My main interest is in the > speech recognition area, realtime recognition of speech units (eg. > phonemes) rather than word comprehension. > > If anyone knows of any useful references, leads or keywords for > searches, I'd be glad to hear from you. This is way out of my field, but I've been hearing some exciting things about a voice recognintion engine being developed at Carnegie Mellon University in Pittsburgh. You might try asking on comp.sys.next.software since I know it's been getting a lot of attention in the NeXT community. Also, I'm going out to NeXTWorld expo next week, and if you ping me in a week or so, I can tell you what I find out about it there. > -- > ----------------val@marsh.cs.curtin.edu.au-------------- /\_/\ > You can't know what love isn't... FOXY / O O \ > ...until you know what love is. LADY \ \./ / > ----------------------------------------My other car is a broom! BTW -Like your .sig! How about sending along a .tiff of yourself sometime? -- --------------------------------------------------------------- John C. Randolph Information Technology Solutions, 400 W. Erie, Chicago, IL 60610 Phone: (312) 587-2000 jcr@its.com, john@its.com, john_randolph@its.com Pick an alias, any alias. NeXTMail invited! Disclaimer: I take full responsibility for all opinions expressed anywhere, at any time, by anyone. ------------------------------------------------------------------ From luosu@ee.su.oz.au Mon May 24 22:18:44 1993 Received: from brutus (brutus.ee.su.OZ.AU) by marsh.cs.curtin.edu.au with SMTP id AA26171 (5.65c/IDA-1.4.4 for ); Mon, 24 May 1993 10:18:48 +0800 Received: from parlo.ee.su.oz.au by brutus (5.61/1.34) id AA23451; Mon, 24 May 1993 12:18:44 +1000 Date: Mon, 24 May 1993 12:18:44 +1000 From: luosu@ee.su.oz.au (Luo Su Huai) Message-Id: <9305240218.AA23451@brutus> To: val@cs.curtin.edu.au Status: RO Hi Dear Valerie Hall, I read your mail from the news net several days ago and have browsed through the articles you kindly offered. My name is Suhuai LUO, a Ph.D student in Elec. Eng. Depart. of Sydney Uni. I've done my Ph.D. work for more than one year and the topic is about synthesis mouth( maybe further to the whole face) through speech signal analysis. So you'll see how glad I am now to be able to talk to a person who has done so much in speech driven facial animation. I have not kept tight contact with researchers in this area before, so I hope I can make great benefit through our following correspondence, if you agree. As to the speech driven facial animation topic, my aim is to vividly derive the whole series of mouth(face) movement images based on a segment of continuous speech given. My main interest is in pattern recognition: both in speech and in image. For image aspect, try to derive mouth(face) movement paremeters from real face image series analysis; and for speech aspect, try to recognize speech in the units called visemes which are the phoneme categories that can be visually discriminated. I haven't any experience in animation, and the graphics facilities avaliable to me here is not enough, so I am not sure how far I can go in animation and what far is necessary to fulfil the research. I hope we can keep in touch in the future and both of us can benefit from this connection. Anyway, we are not far apart. Please feel free to talk. Thanks. Suhuai Luo Speech Technology Research Group Department of Electrical Eng. The University of Sydney NSW 2006 Australia Tel. 02-6924509(w) Fax. 02-6923847 luosu@brutus.ee.su.oz.au