5-1345msg1.txt

贝叶斯学习算法源码

dear colleague , i be really grateful for your assistance and your information , which be present bellow : a ) for dos / window 1 . speech viewer ( with no more information either about the manufacturer , the price and any comment about its function ) . 2 . cecil ( computerise extraction of component of intonation and language ) price : around $ 300 8 bite sample @ 8 , 13 , or 19 . 5 khz , to dos pc memory , vium parallel port . software extraction and display ( ega or vga ) of various feature , e . g . , waveform raw f0 smooth f0 etc . has built-in ipa font for transcription / label of analyse signal . address : " jaars ( international computer service , box 248 jaars road , waxhaw , nc 28173 , ( usa ? ) ( 704 ) 843-6151 , fax : ( 704 ) 843-6200 ) " ( kimberly soto ) . ( it seem that cecil be hardware of the follow package ) 3 . sil speech analysis system a package that " do a very good job of track and display f0 . it will also display the waveform and / or amplitude contour simultaneously and can do some rudimentary spectral analysis as well . the total cost for the hardware interface and software be about $ 300 . the hardware component consist of a small box that connect to the parallel port of the computer and have input jack for a microphone or tape recorder . the version run under ms-dos , although i ' m tell a window version be either available now or will be soon . the address to order from be : international computer service attn : customer service box 248 waxhaw , nc 28173-0248 usa phone : 704-843 - 6257 they also list an e-mail address : icsust @ 1 . jaar . sil . org . i have not find this address to work in the past " . ( rod casalus ) 4 . csre ( canadian speech research environment ) . the program be develop by don jamieson and other in western ontario . it need a 386 pc or more . the program cost $ 400 ( in 1993 ) and the manufacturer recommend an ariel board as hardware ( which cost around 2 - 3 thousand dollar in 1993 ) . " address : avaaz innovation inc . po box 8040 wonderland rd . north london ontario n6g 2b0 canada tel ( 519 ) 472-7944 fax : ( 519 ) 472-7819 " ( franke ingolf ) 5 . dsp ( digital signal process ) the manufacturer be : ariel corporation phone : ( 908 ) 249-2900 433 river road fax : ( 908 ) 249-2123 highland park , nj 08904 dsp bbs : ( 908 ) 249-2124 ( 300-9600 bp ) ( i believe in usa ) in addition , they offer the speechstation , a complete speech-synthesis package . 6 . speechstation ( sencimetric ) ( with no more information either about the manufacturer - which could be the ariel corporation - , the price and any comment about its function ) . 7 . cspeech " for a dos environment , cspeech do a great job of display a waveform , a fundamental frequency contour , and an amplitude contour , ( as well as other analysis , include a spectrogram ) on one screen . for further information about cspeech , contact : paul milenkovic dept . of electrical & computer engineer university of wisconsin - madison madison , wi 53706 u . s . a . milenkovic @ engr . wisc . edu " ( charle read ) 8 . kay elemetric ' computer speech lab ( csl ) " very few of the commercial pitch tracer be good with noisy recording . kay claim to have a new super-robust system which work with their cls work station . " ( linda shockey ) " this set-up must cost around $ 3500 - $ 5000 without the dos machine itself . " ( alex franci ) " the best pc system for general purpose speech analysis , include spectrogram linear prediction analysis and all kind of other thing , be the csl system from kay . it be a lot more expensive , but be really quite sophisticate and very impressive . strangely , though , the f0 track be not very dependable - there be sometime double and halve , and quite often you need to fiddle with the parameter to get anything at all usable . " ( david deterd ) " address : kay elemetic corp . 12 maple avenue po box 2025 pine brook , nj 07058-2025 usa tel : ( 201 ) 227-2000 fax : ( 201 ) 227-7760 " ( franke ingolf ) 9 . loughborough sound image speech work station this program run on a pc at ( or 286 / 386 base compatible ) , with 640k ram , ega / vga graphic , microsoft mouse ( or compatible ) , hard disk ( 40 mb recommend ) , ram disk ( require for stereo record or fast sample rate ) , dos version 3 . 0 or greater . the lsi speech workstation can display the signal in a variety of way , include black and white or full-colour spectrogram , waveform , spectral slice ( cross-section through a spectrogram which be display horizontally across the screen ) . . . all of them be reasonably fast , especially on a 386 pc . a wide range of bandwidth be available for the spectrogram and the spectral slice , and the waveform can be scale . . several of these can be display at the same time by split the screen . the screen can also be split to accommodate part of two separate recording . the analog card supply with the speech workstation have two input channel each of which can be connect to either a microphone or line output . two marker be available , which allow you to pecut and past , copy etc . it be possible to play only mark section of the signal on the screen . the maximum length of analyse speech be 3 - 5 minute . the manufacturer be : loughborough sound image limit the technology centre epinal way loughborough england le11 0qe telephone : ( 0509 ) 231 843 telex : 34 1409 lufbra g fax : ( 0509 ) 262 433 10 . sfs " i ' ll include the whole readme file : speech filing system computer tool for speech research department of phonetic and linguistic university college london introduction sfs provide a compute environment for conduct research into the nature of speech . it comprise software tool , file and datum format , subroutine library , graphic , standard and special program language . it perform standard operation such as acquisition , replay , waveform edit and label , spectrographic and formant analysis and fundamental frequency estimation . it run under unix and dos environment and be currently run on sun , hewlett - packard , masscomp and 486pc . sfs be copyright university college london , but be currently supply free of charge to research establishment for non-profit use . sfs be supply as be with no warranty or support . feature operate environment : unix , protect - mode dos ( with gnu compiler ) support data acquisition / replay : masscomp : ad12f , da08 sun : sparc - 2 8 - bite , sparc-10 16 - bite ibm-pc : data translation 2811 , pclx , ucl parallel printer dac ( sfs support network replay from unix to pc ) support graphic device : masscomp : 6 - plane colour graphic sun : sparc - 2 monochrome console , suntool sun , hp : x - window pc : vga and svga ( sfs support network graphic from unix to pc ) epson 24 - bite dot matrix kyocera laser printer postscript laser printer wordperfect graphic file output utility : create sfs file , list sfs file , display / print sfs file , copy / link / remove item in sfs file , dump contents of sfs file . analysis program : acquisition and replay , waveform process , laryngographic process , fundamental frequency estimation ( from sp or from lx ) , formant frequency estimation , formant synthesis , spectrographic analysis , lpc analysis / synthesis , filterbank analysis / synthesis , psola prosody manipulation . file format : import from text , binary and ils file ; save multiple datum item in sfs file and compare ; standard format for speech , lx , tx , fx , annotation , synthesizer datum , spectrum , spectrogram , lpc coefficient , parameter track , etc ; export to binary , text , ils , htk , etc ; process history maintain in file . subroutine library : support sfs file i / o and dynamic memory allocation for datum set ; matrix operation ; device-independent graphic . special purpose language : sml speech measurement language - interpret language for measure datum in sfs file ; spc speech pascal - compile language for waveform manipulation and analysis ; c-span - compile language for synthetic speech stimulus generation . source sfs be available by anonymous ftp from : pitch . phon . ucl . ac . uk in the directory / pub / sf ( from august 1993 ) . the readme file give current version information . remember that we be unable to service request for support on this software . bug fix only may be send to sf @ phonetics . ucl . ac . uk ; request for help may be ignore . acknowledgement sfs have be develop from software write during the spar alvey project involve gec , imperial college london , university college london and leed university . the software that be distribute contain only the ucl contribution to that project . additional , compatible software may be available from these partner or from other current user of sfs , for example at york university . please contact mark huckvale for further information about ownership and other available software . mark huckvale university college london gower street london wc1e 6bt sfs @ phonetics . ucl . ac . uk " ( hann pirker ) 11 . spectro 3000 " 2 channel signal analyser ( separate devise ) this analyser have the best pitch technique ( sift and cepstrum ) i have ever see - but be very expensive ( about 50 . 0 dm ) . address : medav digitale signalverarbeitung gmbh graefenberger strasse 34 d-91080 uttenreuth germany " ( franke ingolf ) b ) for mac 1 . voice navigator . the only thing that i learn about this software ( ? ) be the name - only - of the manufacturer , i . e . articulate system . 2 . macspeech lab the manufacturer be : gw instrument , 35 medford st . , somerville , ma 1123 ( canada ? ) ( 617 ) 625-4096 ( 617 ) 625-1322 ( fax ) 3 . macrecorder ( no information about it ) 4 . dsp ( the same as for dos / window ) 5 . signalyse ( version 3 . 0 ) " signalyze ( tm ) 3 . 0 be an integrate speech signal analysis application for the macintosh . it do signal edit and direct signal i / o to / from a number of device . version 3 . 0 have a user-friendly multi-level label feature : each label be code for a linguistic level ( e . g . , segment , syllable , etc . ) . level name be determine by the user and be color-code . also new in version 3 . 0 : speech slow-down and speed-up ( up to five time ) , color / grayscale spectrogram right with signal , av - macintosh support , easy vertical zoom , and more . signalyze have a large number of spectral analysis tool : spectrogram ( b / w , 16 and 256 color / gray ) , cepstrogram , cone kernel , lpc - gram , fft spectrum and cepstra , and lpc spectrum . also include be statistics , db measurement , interpolate signal resample , transformation , envelope , zero passage , and filter . the manual be 224 page , the on-board contextual help be in english , french and german , and the whole interface be switchable to english , french and german . the program be about 980 k at the present . it run on any mac from the macplus on up ( 4 mb and hard disk require ) . price effective january 31 , 1994 : individual license : $ 350 . departmental license : $ 750 . organizational license : $ 1250 extra manual : $ 25 per manual . ship cost : 1 . u . s . , canada and europe : $ 10 priority / air mail 2 . rest of the world : $ 20 priority / air mail 3 . 3 - day ship anywhere in the world : $ 50 " also , " here be some detail on the new label facility in signalyze ( version 3 . 0 , macintosh - specific software ) . this may be of use to people work in prosody . levels the signalyze label operation work by level . for each label , you specify a level , such as " segment " , " syllable " , " phrase " etc . each level have its own label color and its own , user-definable name . label be mark for their level by a number place in front of the label name ( e . g . , " 4 : its " , means that the label " its " be mark for the fourth level ) . what gets labeled you can label either point in the signal or select portion of the signal . the label for select portion be place at the center of the selection and be mark by angular bracket ( " ) . . . ( " ) . re-editing , adding and deleting labels label can be re-edit , new label can be add anywhere in the signal , and label can be delete individually or for an entire signal . reproduce labeled segment while the label be open for edit , you can play the select portion of the signal by do command - y . if you ' re label a selection , and if you ' ve set the audio to play signal selection , you ' ll hear the segment which you ' re about to label . align labels you can choose any of nine different vertical position for place label . when you close a label with a click in the close rectangle or with return , the label automatically " snap " to the nearest standard position . save labels the label information be save as a tab - delimit text file with the extension " . lbl " . it be store in the same folder as the signal file . label format the label format be in the public domain and be fully document . it be available on the signalyze server ( see below ) . transparent saving and opening open file operation on signal file that have accompany " . lbl " text file cause the label information to be read into signalyze . save file operation on signal with label information automatically save the label in the same folder as the signal . save file operation for signal without label information erase whatever label file may have exist in the folder . label in phonetics you can use any phonetic or normal font in the label . however for phonetics , it be recommend stick to sigpalfont ( the shareware font supply with signalyze ) . sigpalfont preserve the number and angular bracket you need to indicate label level , which be usually not possible with other phonetic font . search by label or by level you can search for label use either a give label level or the label 's name . you can specify two separate search pattern . switch between the two pattern with the shift-lock key . more info and demo information on signalyze version 3 . 0 be available as follow : by ftp : ftp macfl4082 . unil . ch or ftp 130 . 223 . 104 . 31 login anonymous by gopher server : name of machine : gopher . unil . ch , find " europe " and " switzerland " , select " university of lausanne " , select " autr gopher de l3unil " , select " faculte de lettr " , select " laboratoire d3analyse informatique de la parole ( laip ) " , select " speech analysis and speech synthesis " , select " signalyze " prof . eric keller ( new email address : eric . keller @ imm . unil . ch ) laboratoire d ' analyse / / ^ | | | | = = \ \ informatique de la parole ( laip ) / / / \ | | | | | | lettr , universite de lausanne / / / / \ \ | | | | = = / / ch-1015 lausanne , switzerland / / / / = = = \ \ | | | | fax + 41 21 6924639 / + 41 21 692 4510 / / = = = = / / \ \ | | | | " . 6 . ucla - uppsalum analysis package ( to run on macrecorder file ) " write to : software manager , phonetic lab , linguistic dept , ucla , lo angele , ca 90024-1543 for our order form . it 's $ 5 for just this disk . " ( peter ladefoge ) 7 . gw instrument soundscope ( formerly macspeechlab ii ) ( mac ) " more able to be tailor to individual use than csl , but also a little clumsier and slower than csl . i think it run about $ 3000 . " ( alex franci ) c ) for unix 1 . xwave + it cost around $ 5000 and the manufacturer be : " entropic research laboratory , inc . 600 pennsylvanium ave . , se washington dc 20003 usa " ( franke ingolf ) 2 . dsp ( it have the same specification with the sfs that have be present above , about dos / window ) . 3 . sfs ( it have the same specification with the sfs that have be present above , about dos / window ) . 4 . digital ear no more information , only the name of the manufacturer , i . e . metaresearch 5 . ogi speech " this be free ! ! ! i have never use it professionally , but while work at lo alamo nat ' l . lab i do get a chance to play with it . it come with a good manual , and work with a number of different sound-file type , and can be configure for a number of different platform ( sun & sgi ) . . . . it need some additional hardware . " ( alex franci ) . 6 . entropic sps software that run through the xwave + package . i do not know its prise ; also , i do not understand if it be a latest software improvement of xwave + . " address : entropic research laboratory , inc . 600 pennsylvanium ave . , se washington dc 20003 usa " ( franke ingolf ) additional information : read , buder & kent . ( 1992 ) ' speech analysis system : an evaluation ' , journal of speech and hear research , 35 , 314-332 . p . c . bagshaw , s . m . hiller & m . a . jack , ( 1993 ) , ' enhance pitch track and the process of f0 contour for computer aid intonation teach ' , proc . 3rd european conference on speech communication and technology , pp 1003 - 6 finally , " natural language software registry the natural language software registry be a catalogue of software implement core natural language process technique , whether available on a commercial or non commercial basis . the current version include + speech signal processor , such as the computerise speech lab ( kay electronic ) + morphological analyser , such as pc-kimmo ( summer institute for linguistic ) + parser , such as alveytool ( university of edinburgh ) + knowledge representation system , such as rhet ( university of rochester ) + multicomponent system , such as elu ( issco ) , penman ( isi ) , pundit ( unisys ) , sneps ( suny buffalo ) , + application program ( misc . ) this document be available on-line vium anonymous ftp to ftp . dfkus . uni-sb . de ( directory : registry ) , by email to registry @ dfkus . uni-sb . de , and by physical mail to the address below . if you have develop a piece of software for natural language process that other researcher may find useful , you can include it by return the description form below . if you be interest in the preliminary draft of the registry , do not hesitate to drop us an email message and we will be happy to send it to you . " ( jane edward )