GB2416874A - A translation and communications system - Google Patents

Abstract

A translation and communications system includes voice recognition means and a lexicon which includes sentences, phrases and/or words in a plurality of languages. A code is allocated to each phrase, sentence or word such that a plurality of phrases, sentences or words, each in a particular language but all having the same meaning, share a common code. When transmitting between a sender and receiver in a universal communications system the system only needs to transmit the code rather than the entire sentence/phrase and the receiver's system can replay the phrase or sentence in the required language using the code and the lexicon. The voice recognition means may include the ability to adjust a users's voice waveform to a fixed constant fundamental frequency. The voice pattern can be modified by incorporating two microphones, one of which is positioned in the region of the larynx and the other positioned near the mouth.

Description

24 1 6874 A translation and transmission System A Universe Communication

System

Introduction

111 a world which is getting closer together due to fast transport and elech-onic com.n.lu".icion hetree!1 distant pa - s fifths p! filet. one ofthe lent retprohiems to be solved is how to enable people with many din erse languages to communicate win each other as easily as among themselves.

At presents this problems is not vet too difficulhvllile the main industrial trading nations haste a reasonable compromise in the English Language and only simple sentences phrases are r equired if a foreign language is required for toluism.

However as the Far East and Easten1 Elupean coulh-ies etc. become economically important. it will he imperative to either haste a Universal Language or a 1anslahon system capable of operating in real time. Additionally, within a vast counh y that contains many etiluic sublanguages and diaiecrs, it is diilicwT Ior people to communicate Witil each other either by speech or text.

Problems.

If a person wishes to talk to another person who speaks a different language, he either has to be able to speak the other language or learn a common language such as Esperanto but these hvo options require the person to learn a completely new language, or speak in his own tongue and expect the other person to understand him. It is possible to speak tlu ough an interpreter but this is disjointed, takes time and it is difficult to know if your thoughts are correctly translated.

Kecently with the advent of i:Omputers Witil large processing capacity. even it it were possible to translate in real time in only one foreign language, when it is required to do so lo 111ly IIgUC'7 ii UCCO111t'7 loo costly All catty PHI SO'i1 Hi illiS iilll7 ii will 14Uil very powe ful computer and also the time taken to progmme them will be too large.

- ..is secr.a. io holds v' both panics in a con. c. satinc,vrecr. ,- o di=.crcnt l.gc speakers.

/foth*r problem. is the At ofthe 'abound- width" requiters to c. .. the increased irormation on an increasingly overstretched telephone system. even using digital tT=lSmiSSiMl.

Ale Invention would go some way to alleviating these problems arid has the advantage in that tile person does not has e to learn a nerve languages Rich usually tastes years to achieve and for some people it is irossible'but he must only teach his conpute.- sentences' phrases alxl specialist words inhis OAI langlng;e Old Enrich. call he Basils done incrementally over aperiod ol time.

Background to the Invefioll.

Ihe basis ofthe Invention is to first create a list of Sentences. Phrases and words in one i-eUoH-llijal-lt til-ouae bliti-i as Englibil iliac COill eys the esbei-le of uijivrsni conversational usage to create a colon Lexicon. and match it to the equivalent Sentences, Ph"-asec; find herds in ^1!a.O. ages erthe vrld such that tl.e se.eaninO is faithfully conveyed to all users ofthe Lexicon.

rh!r it o the cou'se of a cotn7erntion teen halo peo,r!!e the common nhietive is f r both persons to understand what is said and even though the person can speak another language he mav not be understood because of his accent. For instance far eastern oconIe because oftheir language cannot pronounce certain words or letters as there is no need for them to cultivate certain sounds, making it difficult to be understood by a foreign person.

The Invention may be applied to all types of communication, but to explain the principles. we shall concentrate on languages that can be transmitted by tided wires or optical cable etc. such as the telephone or computer networks etc. or by radio waves illCIU(liIIg b'Cl;iiC COIIIIIIliCiiOIlS.

When people converse with each other' they compose set sentences in discrete blocks co on to their l.guc.d cult". c to COn'VC Y the sscncc of t.hc co creation, c en though certain lacy words of the dictionaries may be substituted to change the meaning coded. A sin1n!e fogy is the to''risthrce book. It is also Mite common "o.l.r2d lays to transpose certain vell-hlown words or phrases from one language untranslated in another language such as trade names like Coco Cola and facilities such Do Toilet and Exit etc. W1len speaking any language, the meaning is composed as a sentence but the order ofthe words in tile sentence is not always in tile same sequence, tor instance, the verb placed at the end ofthe sentence in Clermal1 as opposed to the English order or many words may love lo be used ill Alla IcHlUcU ivy il-cHlS;cliC alla WUl-U ill cUlUillUl- ;cDl9Uc. A irm;cu- would recognise this and transmit it in the proper order that can be recognised by both pc"-tics.

lit all the sentences and ply a.ses used in speech common to all languages were compiled into a T exicon that may or mall, not include Hi! the specialist Ids it! the ditiona'-y that may be used such as by doctors or scientists, then it is possible teach this lexicon into a voice r ecoonition system that can recognize speech in the form of sentences and phrases but need riot r expire it to be able to <cognise the meaning as the meaning is already in the sentence.

The sentences in the Lexicon in one language would be diercnt front one another in the Equal words but the meaning of each senlence.o''id}'e the satire tor adi immures. In, certain cases 'l someone rectiv trmsiateC the orcl ecluiYalents in anotlle'- ianguage the other person' may not unclerstanci what be wants to sail.

1nis -steno would recluirc a IDuch i onoer airung rime on!ne computer. out w-oulci no! require the more c omplicatecl pr ogramrre net essay for uroi ri recoition anti interpretation, if it also had to guess the meaning ofhat Bear being said. At least it Florid be muci1 easier than iemning a language anti much less so for many languages. it is envisaged that the indiN-idiial will be costa-lv adding need material to his voice il'clillill PI v-cHlD!l 1O CVUIIilllY IIL' \Vill cUSO tLO111t C1-> plIVIIL'iffi ill ills SiVlt Oi ills composition of the sentences consistent with his specific Lexicon.

TO odds to '-ecu "A to dot- ;i'.h tilt c. A.- ciicuiiii Ice '" ClltC'. i.. +illl, trio computer up-dates the latest sentences used in its database and adds ens necessary i'{nrn3tion s''rh:' rh:in2:, c in the.:n.ice rtte:rl.

All clew additions to the Lexicon in any language must be common to all la'ua,es and Allen new additions become necessav there must be a central co-or&nating entity that adds the new material that is compatible to all languages.

Ale invention now calls for each ofthe individual sentences and phrases in all the common Lexicons of eacli language that Han the same meaning, to be assigned the same Code or member or other identifying marker tl is common for all the languages. The Code etc. is preferably the only information transmitted and may be in digital or anaiogue form, and takes up a fraction of the "Eland width" compared to the equivalent N70ice ildllbilliHbivUl Ui ills VvilUit StilillL' Wiv^till 1111-t UL'vilvilliL'i Use Of ills IllUt;iUIIL TURF is also provision for the code to be encrypted by mutual consent of the parties in contact 4_..;6,. Am_ 'A AL" .U111,. 1C=IV;I..

It is also necessary to record the actual voice sample ofthe sentence phrase or word as olden hit the sneaker Unto his opal comtuter'lerllorv during tile teaching process.'nr' to be able to use its Digital Code to play back the same recorded sentence ofthe speaker's pre-recorded Mice.

The end product results in the ability ofthe Digital Code to trigger its equivalent sentence in any language simultaneously using the receiving party's OWFi voice, including his accent and other attributes. This makes it easier for the conversation to be better understood as it is sometimes possible that the transmission medium has static and is not perfect for normal voice communication, since the receiving party hews his own voice recording that is triggered by the digital Code transmitted by the other party and which is IWi tiiy L'0IIUpiO ibiUl'iU U='iil il-liHSHii It will now be possible for someone to speak in one language and be understood in all the Big "C5,a 0 the vie. ! i..."ltc" is 0 as i y Ad all I.l ic v, o. i "one i n c as I I scp ill c" c c" lucs,c by a speaker or speakers each taking on the task of teaching and recording all the corn sentences eta ofthe Leicn. n in his nylon!.guage together with the equi.'ent code that is used to communicate with the other language speakers.

Even when real time conversations tulle place that e will be little observed time discrepancy in the conversation eNren ii the digital number is sent in packets" via mIy medical including the Intemet instead of a moi e expensive dedicated telephone line. This packet system is ideals especially when there is disruption in the transmission due to heavy traffics by sending back-ul packets e. g. by way of the Interpret and the r oute changes atomically it there is congestion or break in She iinl;. Icleully a fast link is rc.ircd such cats cable o;- optical systems includir',B;-oLLL113<ttld to the T:.re:-ner for instant l. 'OlilIIIUlliL'[UiUII, cl5' i;lrl-t: L'UtIIt! Us tIIOY'ill,9 iililt [imply; ill t11U L'UIIVtI.S<IliUII Out is ills slow colorer cables ofthe local loop and the intein'ional cables i J, . _ -. i... .. . . . . a. 1.'1. i. r inch lc;l..t i; the in. c illit!li ill tout a;. t.;c.. t!; SILO; pal 02; 11 bile l [ O;Icc t.lc has e; ofnner] ark trainer! their voice reco, itic1l system using the common I.e.xicon :.

sentences and phrases, can now communicate with each other using any com!llunicatiMls mediums In use today.

One drawback is that the personal touch of hearing the actual voice ofthe other part is USClli UUi ills iUlI iUUL'IleS ilill elk ill e.T-tSSill ite; IgUe iS pI-eSCIli. EMU cues' ii the other speaker call actually speakyour language it is often difficult to understand him L;c to his acecr.t such - a C'hi.,csc spc arc;- SpC&r.ilg i.' E;ilis'i. Talc play balk Voice may also be pro'3ramned electronically with an accent of the other foreign speaker or even from male to female to lessen the impact.

It is also possible to trigger the Voice of an ideal speaker of ens one language in the case of making speeches or lectures or for delivering the Newscast to an audience in their own language delivered lay a foreign speal-er uprose voice and computer triggers the digital Code. In this case either the code or the already translated voice can be transmitted.

Another application is that the transmitted Digital Code can easily he used to simplify the print out of meaningful text in the language required that is now done by using a commercial voice recognition dictation appiicatioa In this case the;erences and Phrases are typed out during the \70ice recognition teaching session or it can be done sepiu- Hie;) by the UiSiliUUiUl- of the L'Cllil-fiSCULe.XiL'OII SySielil tilt ills SClliCIILtS lilid Phrases and assigning the related Digital code that is used to Rigger the typing and posting the data or presenting it for doYncadir.g on the intc'..ct by clicks rcgistc. cd as users ofthe System to load into their memory system. This print-out man have to be edited by the rerifpient m-anl-lally or'roorllnme1 to link the sentences into a literacy stV, le for proper reading.

The Invention.

A Universal Voice or Text Communications System comprises in combination. The lexicon' the code, 1 he Voice Recognition ysten and The Method of transmission of the code.

The Lexicon.

The Lexicon is made up of a compilation of all the possible sentences and phrases that are required in a con.rersatio" in hero or mare ifferenf!n, guages, and is upped continuously as new sentences come into use. The composition ofthese sentences is based on all the Variations possible within the ocabula'.,' and dictionary of the language Ed the Lexicon expands as new phrases are acldecl E,rentuallv the whole system takes on a language style of its oven as the leaning curve creates its oven distinctive form.

It is essential that the translation from one language to mother must be by means of whole sentences or parts of sentences phrases or words that have the same meaning although specialist words may be used as required using its own code, to indicate certain things or technical terms. To achieve a high led el of co-orclination between the languages l cl e;iuT us I- I.it:J'! I!!U<'' r!,T-.;fl-- Ale. mu., allele' lill=,uisl ill bill i'= it:; i, gi e the best matching sentence Ed they 11lUSt co-operate with othel- specialist liquids i a, c. cti;., the tic Pi c.1 1 am. a a - . The Code The sentences phrases told specialist Voids within the Lexicon are each allocated a 3UL'iliL' .UtIC ills is ills U!lly ilOUI Illil311 il-rliliii UiWII illt StilElUl- mid l'ECtiVI.

Ale Code is preferably in digital form t}lat can be easily sent through the normal col.lmr.icativi. cl.an".ci o.- :h.;;gih the Tnt.t a. praC'-CtS tt'ic'i Caui imp. the Burl ions routing options available at any one time.

The nitrite! Pouffe ca' be t.r I, bend i To Awoke fO''7 or.7ie-.7e a {or Cm_e transmissions.

It may be that during a conversation we utter a sentence that is not within the Lexicon and therefore no Digital Code would have been allocated In such an instance the voice recognition system would prompt the speaker to repeat the meaning of his utterance in a different form of words s a sentence. It nary be that the Computer or even a mobile phone will be able to brin up on screen a drop-don list of sentences to choose Born based on words in the rejected sentence it can recognize and searches for alternatives. At an early stage in the training process many such mistakes would happen until we VIIiU;y L'Ulli'lil it ills LXjL'UII; HIIgUt Si)';C ills WEDS UligillH'ly igili is ills computer or it may be that eventually with close co-operatior our mistakes may be incG'yoraicd ithe u." -Sal Lcxico.l. Ithe.-occss oftilc the T.'iCo.l 'will cerise a language system of its own which is common to all people.

Ale T lexicon roll! be ndatPfl by revising or ad1in m ore sentences, neuroses anal worrls that are necessary to improve the system and it is therefore necessary to centralise this function. This can be done by anersletter system that can be sent out to subscribers or dow.nloadir through the hlternet. The recipient is then required to teach his recognition system while allocating the corresponding Digital Code and recording his voice as he teaches the system and which is triggered by the Code when receiving telephone conversations or announcements etc. The \70ice Recognition System There are many automatic speech recognition technologies that convert speech into the written form in real time. 1HM and Dragon are two examples on the market today. They all use basically the same technology with some slight Variations to achieve a commercial IY-_Ii-. Ills H1111 iS is Allele yl-SUII SyttL'II ii is Utile SyUhUll' is Ut CUING U1 itt; into text with a high degree of success and therefore this requires Avery con'licated o..."lc, y! O....'C.

All these systems hare to be able to overcome the pa oblems of Data r ate' ethnic Speaker a iai!iu. same nea.-.er striations a cl Tonain dependence a! .! they anally e il e 'ioital iormEition of the spoken word and print out the equip Ede}lt text in the order that it is spoken. It is le.ret.!-e tin iteA. to one lan'uae.

The voice recognition propane samples the speakers analogue oice as es in dig;iml form as a rumba-. many times per second in order lo ninimise the Data r ate tu the computer. This cuts down the size ot cor'iti'c pow r to a r ea.coMble level using nexens!ve hardware. is

Domain dependence is covered to a great extent in the Invention as it uses ilole sentences and phrases that cone the whole meaning.

Ethnic speaker variability is also a problem since vvithin the language there are different t'L'ClliS Flu Slllml dierellLeS betweell m-r of ills CUUllil-. 111iS ('1 llbillty be Solved during the teaching prom am for speaker dependent systems but not easily for speaker iepr.dent systcs.

A big problem is how to sole e the \ ariation in the same utterance from the same speaker because every. ,tern.ce is unitize.ith regard to pitch and speed of Eve. etc. ..1nen a speaker repeats a word or sentence after a long elapsed period oftime, there will be slight differences in the fundamental frequencies due to emphases or stress etc and it is essential to find a solution to this prollen.

The Invention proposes' by w;y of example, tore preferred solution to this last problem that are used to give an idea ofthe scope of the invention.

When we speak, air from the lungs pass over the vocal i-olds' in the area of the Adams apple, that vibrate to produce the Fundamental frequency of that personas voice at the time it is spoken. This vibration is modified as it passes through the throat nasal passages i:UIt; IllUUill, illL'; UUill ills 1UIIgUt' pIUUUL'illp lmlllollics'cuiU OillUl- ViOlHiiOllS illcU 111 Up the speech sound If Eve speak normally using old vocal folds or whisper in which the folds a. c silc; t o. . c1.im.,c."hc 'vV do..c salc physical Tic. ations occu. in the treat mouth and nasal passages for the same words Die hrm-on-icc or e Annie nnlltinlec nfthe fi!ndment^! f eollency..nr3. rue user by speech scientists to analyse speech. These harmonics can be observ ed when using a sound spectrogram that produces a two dimensional picture ofthe harmonics called Formants.

Rile sound emitted by a person is in the form of a complex wave pattern that comprises all the frequencies of these harmonics and it is this sound that has to be sampled and analysed by the computer programme to identify the spoken words. Mile speaker may also on different occasions emphasise certain words by raising or lowering the pitch of his voice thus varying the fundamental h equency and hence the her noilics.

This variation makes it more difficult to carry out e.g. a Fourier analysis on the complex VVrVe E,utielll to OI-iV ills Silly Sills waives SSOL'iiCU \\'iUI C='11 11 1llOlli 11lUlltiOlltU above.

Thovvwcr we cam. be.c to spa in a.ooto T oicc' iwhich the l"&lll.ctal frequency remains relcAdiYely constant' the Formants being harmonics inherent in the Broken so-d rare easie to nwyse the cognate wAreform is constant fo - each sentence, phrase or word spoken malting sampling by the analogue to digital converter iAI:Ci more reproducible. Onlv the intervals remain which does not pose any problem to overco He in the programme.

To achieve this monotone v oice electronically the Intention uses essentially two microphones that are fed into any suitable electronic system that is described below. One microphone is placed as normal in fiont of the mouth and the other is placed over the region of tie Cocci oicis in the throat redone to pick up Quit' the fulciaental frequency.

This I;.ndental Sequencer short. the vocal solely mic; ophone is p! Jcessed to filter of thil-cil!OUS SQUIDS, HI is tSSIRi;lN pt8- Sole VV. llliS tVclN IS IllliiLilLi 1 ChallgeS in 6'eqUellCV ctHd speeci of delis ew and a feedback system is used to track this c q;; c rl. . a r I a. a I A I. I; i i.; a 1 LL. t c t c r, / . I I v p d t c t i s Vised in r aclio r eceis ers. Fin. 1 shows a Italic PLL circuit. f,

This fi'ndnental fi-equency change is analysed and an adjustment is constantly made with reterece to a pi-deterfnined constant frequency for that person. The acijustment feed back current from the PLL is used to modify the Fundafnental fi-equency of the IllUUill IlliL'155flVIIt lU1 Civilly ii ill filly Wiill ills' --iCl-lfilt'U L'IlSicUli FUIIfflIli'l Frequency for that person.

It i pa S i Al C t 0 phi a ic a, I i sit h o; o.q. e the Icu.,.oni ova c Act ' aLtc' the.y flu created and recorded' and hear the same meaning such as a Phone voice sound ten pied At different speeds can be easily Imdersto d.

A monotone fimdamental frequency passing through the mouth and nose passages creates harmonics at the time of speaking. If the fundamental frequency alters, the harmonics will change' but the same mealung received by the listener.

ETsing this PLL technique, it is now possible to examine the speech sound complex waveform for a word or sentence and snatch the profile against many such repeat recordings for an individual in a dependent speaker mode' or from a sample from one group of a population in a indepencient speaker mocie. By analysing the prod ies it is also possible to select the parts oithe profile that is common to the inclividual or group to U-i ill;t ills L'all Ut USED ill ills '1 1-CL'OgiliiiOII 0 11111t lVI CitilUl- IIIU.

Another preferred system using the monotone approach, is to train the recognition system A' pa am. .rl.i n in c fi.li. c;.o c. no I h o. A ng p, of;! c to spc c oh s c ugly (cormparable to sign language for deaf and dumb people) so that the speaker only has to !tter a iinlited seo,.!ence of sol!Ids to c oared his total voire nrc6file anti if accents. flialects and other characteristics of a population can be also programmed in, then a truly foolproof speaker independent system can be devised. This new approach can also deliverer both speech recognition and speaker identification systems that can lead to wide set of applications.

Another preferred method is to use a simple sound profile matching system that accesses the data bank of sentences or phrases to get the best match. It may be also possible to sequence the matching process by accessing the initial words and if a match is found, to proceed with the next word in the sentence etc. until a good match is achieved.

lOillel yllel red Illrthod to plUdU iht IllOllOiOIIt speech is to Use the 1a Used croci of the PLL detector to alter the sampling rate of the Analogue to Digital Converter \A' such th iwhc.l the CC.IC; discs, ..hc S.',li...C i..C.CclS.S vvhicll is analogous to lowering the frequency to the fixed level. This system requires vent accurate S spooling techniques And feedhAr1 accuracy To achieve this goal, two preferred solutions using the feedback from the PLL system, will now be described.

C,ne preferred method for obtaining the monotone voice pattern, is described in Figs o& in which the analopue signal front the mouth microphone is recorded on to an, suitable recording medium such as a continuous loop magnetic the or wire Fig 2, or a recording disk of suitable diameter Fig:. all nulning at constant speed. In common with all the media a movable recorcii head tai cn-iven nv the feed hack Font the ' it, moN,es between pOilltS,X. valid (V) I({ the plclvhk head it,i and tile erase head fry He fixed in pUSiiitHi. ill ills CXcHllp; Ui'ih Disk ills rrcal;ill leek 1 TJiais cU-UIld ills Spilidit (dl of the disk, and is driven b>, a servo meclalisn' (e) that is conll-olleci by the teed bacl; T I. - , . . A A I- A A al A A i r r.

A. rut oli. the I AL A.c. plc''bc' ran, c' ' cr. in.,, All c" 5. zi.;w posh or. d Al lCU aheatl of the r ecorciina head a possible as shown. the hmdamental Sequence in tne -i throat microphone rises in the PLL system, the feed back cu Tent drives the recording Heidi in tile opposite direetiosn to the spin of the cite effectively depositing a voice recording on the disk from the mouth microphone at the constant predetermined iUlidLililrilictl it gUtIICy. 111t l tCUI dry SVU'IU is rilUL'iiN Ply SiTtiCIId out it Gil all ills ark and is picked up by the play hack head as a mono toned voice. This is analogous to the r,. ,. ,. f A 1. _ _1. i Al _ Al : 1 I COCK LO I 11 Hi Call lOp iIOIC I CCO' runt A: ,; . Cl cOl SpCC".

Lfthe fimdamenta1 frequency drops, the recording head moves in the direction ofthe spin Creatively com.rressn=, the recorded sound to omrensAie finer the!o.!erin.g of the frequency. It' the fundamental Eequency is equivalent to the pre-determined frequency then the recording head remains stationary. This monotone recording is picked up as the tape or wim moves or the disk spins, under the play back head and fed to the ARC and thence to the voice recognition system. The recording is then immediately erased as the play back and erase heads are fixed in position and close to each other and the dim is free to continue recording continuously.

s tile recorciing head Tnoves torwarci to compensate for the inereaseci trequeney it is essential that the disk diameter is large enough to aeeorurnodate whole sentences and HlI-citS Ur;Ol-U ii l-r='llUS ins pLIyUciL'k filly erase l;. Al ills gild hi ills selliriltr illeir is a pause for an answer and the recording head moves back to its original start point ahead a. th. p I fly by c. as c heads up cs. the Paw. stcIS to talk Thai n c Busing it to.,v c forward or backward again.

Another- nrefe7=e1 methorl is to. red the voice as emitted anti calve the tpirl-!m hear erase heads to move back and forward in response to the PLL cu rent as in the previous method and is analogous to the grcA"sllophone reference.

Another simple nethod that does not require the throat microphone is to isolate the lowest frequency of avoice microphone and use it as the fundamental Eequency which can lee applied to any of the Above methods.

The utterance that proceeds to the AI)C: to be analysed is the efore essential!, constant for the sa ne speaker whenever he repeats the shine sentence and can be more easily processed by any co nmercial word recognition technology or an even less complicated URIC WlliC'll is pI-cUIHIICO in e.;usively to l giliS SOliCllL'ES Ub'ill iClilp;iUS IIC-LD net-w-orks or any other easily achievable alternative previously developed but was fonuld + A; Art ' AN A; I AS AT ADA At AD AM -A AA+A_ l AD 1 ;JI C1411 UIJ1 YIllI "11 " 11% " ! V - ! 1 AL The monotone voice is not heard by the persons sending or receiving the message since the digital rode only triggers his sp-.en. ord orii"-!ly taught to end stored in th.e memory of his computer and which contains all the variations and emphasis ofthe language.

reglured the system can be used to identify the caller by accessing his voice pattern and matclin.g it with a database in your computer.

Transmission of the Code The Code may be transmitted over ally mediums including Telephone lmdlines or C:)ptical To hi- . . . it T. . Hi. ' , ',, ' , L''ltS'l-U Of 1111 al- U\'t1 U1 111111. 111t AUKS O1 Ulill!tlUllilLlUlltllcl may be encoded for security or sent iii part ers in the Inset into the reCeiNer,.< eo.;;p;.t '- to t-icr th. ..cs. arc.

Since the information sent is small compared to a conversation ad involves intermittent use of the medians ba:nciwidth aloud mnipieng conic be used to eHicientiy use the available space afforded by the medium He 1llveius I edit be uteri - ibeu access Susie- such revile pilulles ul- other similar applications including 3G rechnolo7 preferably using the Internet or other chic ion s;71. 11c sycl list hay C acec SO.G a mill GhO"C sly etch. can.

optionally pick up the throat signal together filth the mouth signal enrich can be processed in the reck pl2^A-ct: system Tree to Howe And he modified Voice signal is fed to the AI}C and the sampled digital data can now be processed on a mini computer or transmitted through a mobile telephone to a central computer that has the speaker s voice training stored after teeing downloaded fit om the speaker's computer. The service provider (ISP) is one ideal place to conduct this service and the memory required is solar to that allocated for web space by the provider. The data is processed at Me IMP and the digital code is sent onward to the receiver's ISP who is similarly programmed to play batik the receiN eris Voice sound that is then transmitted to his mobile pinhole. 1 he COST will be a loch call md Me Internet charges ofthe IT to any pad Of the world.

O1IICC ills IllYelItiUD iS able lo CLU dovvIr un me processing' WUI=U;:U1 iS IIUW' possible in use the released memory using the Inventions he used to create a Speaker Independent choice Scot..iticr, i YDtcm that dock 'tot. cam c Ad. T Iti'J"I pI OSL Cl bat can Chic T C common pro nme for a lan,uage irrespective of dialect or other pronunciation differences by m:alonmatno the smmfis resmelt no From the,nh sisal purple of the month tongue and nose etc.' that are common to most speakers of a populationtogether with the smaller differences such as accents. Such a system can be created in the laboratory based on the data fed in to create a viable Recognition Pro,rane and eliminate the tedious task of individual programming v \ 1.

Claims (1)

1} A [Tniversa1 choice andTe (:ounications System comprises The Lexicon7 1

ne Coc!e7 1 he Voice necognirior. System7 and 1 He IviernoO OI 1 sminmO the Code.

f A lo' Ilv' sat Mice We of. - _illviU" Y.mI ill =1 is in vvll'h 1 "e T Beacon is made of Revile sentences7 Phrases7 and Words that are constructed Mom the.r!hote mnn=, of Al! rrmmn 'TernreS thy ran he wed m human communication between any language speaker in normal conversanon or in making a speech or announcements.

A IJniversal Voice and Ted Communications System as in Claim 1 and which the meaning ofthe common utterance in one language is translated ill- into sentences or phrases in any- other fan= - e regardless of its length or complexity of anslariom l. -i n'sa, Dice and.' e-= it..'l"rf.S,re as ill lams 1, d Ah the sentences and Phases in the Lexicon May compose the main stem of . T1 1 If 1 1,1, ,1 1 1 LÌt -11!}!C Fl!-tilr-t Em" I!t: {Ct\' 'UtF] I'd = item If m-e You separately to the Lexicon. rim \

A Universal N7oice and Text Corinnunicatioils System as in Claims 1 23 and 4 in ceilidh the indii;iduai whole Sentenced hrac;es and individual hey words are each allocated a separate digital code or identity code that can be transmitted in any iUl-ilMi A! plUiUCU; iiUUUgI! , CUlItIllUlI!C - Ui! liOiUlIt Ui-SitIl! HMlii UHt person to another person that speaks any other language.

6 A a, ni a. s a I a i T. rim C u ril251 "nit tIti units i i;'t ''' ilt 1 ( Lit'';' 1 ? - 3 ' - c".d in Which each utterance with the same meaning in any language has th same nn, digit! Me that a an triPPer the:o,.,p''t. er iffy reei.er that h-AA been taught if om the common Lexicon, including the senders Computer, to replay the equivalent sentence or phrase in the language ofthe receiver that has been originally taught to and recorded separately into the computers memory.

7} A Universal Voice and Text C4ommunicaticms System as in Claim 1, in which The Voice Recognition System may include any commercial speech recognition system that has beets programmed to recognise the voice Sol the sentence etc. of tile sender and allocates the appropriate digital code that has been programmed into the system A U'liYm-m Culiiluli;aiiv sielll as in V lllilit i LOU '' in VVli:ll U7,;V ills common digital code is transmitted and it is used to trigger the receivers voice dcliv-c'rys the asoc.atcd utt-'.cc that had beep. C\IOUSIY taught to the recipients own computel-.

0N A TT'IivP'sal C'onimunicatinnie Svetem as in Claims 1 Purl in which The Janice Recognition System uses a separate pre-processing means that adjusts his voice waveform to a fixed constant fundamental frequency that is able to use a much simplified progranime especially in speaker independent rejoice Recognition systems and also to identify the speaker at the other end ofthe line.

10) A Universal Communications System as in Claim 9^ in which the Voice pattern is modified by incorporating two microphones, one of which is positioned in the region ofthe iaynx Madam s Apple to pick up the fundamental frequency waveform, and the other is positioned near the mouth to pick up the spoken r VV liVlUIlIL 11}A Universal Communications System as in Claim 10 in which the sound from the Iu-'r.x co...p. iscs r"^,linl Y the funl^r.cr.t: fi qL'cacy in the fo. a;. of sir. csscr.ti PI Y simple sine Alive that is constantly monitory and the deviation ofthis frequency fi om afifeA rredetern!,ned 6'nrlannen! firequenc fit! the speat.-er is determined continuously.

12)A Universal Communications System as in Claim 11 in which the deviation frequency, produces a feed-back current that is used to continuous!, alter the fundamental equency of the complex wave pattern from the mouth microphone that incite its harmonics. to the fixed fundame7lral fi-eque7lcy using any suitable feed-back ratio techniques such as a Phase Locled Loop Detector (PLL) svstein used in radio receipt ers and the modit2ed wa-etorrn is then fed to the ALES that cin.erLs i! tioln the apologue to digital form is Ellis e1 sui UlillilUlliLcUiVIL' S\hlielil cili ill Idilil i O. i i STIR i ill iVIIiLI! V71e preceded means for modify id the Aviate patter n of the mouth microphone is to use the ccd-bac'- ..I.I cut fi G.. .L. "Lit to Iced Lip or slog; don- tl.c sa. lq,li.= frequenf., of the Analogize to I Mitral Otter Aid to adjust the eonplex ll waNe pastern to the equiNralent ofthe fxed findcAullental fiecluency an1 the cligital ntormaton s ec'to tile commerccn &'oice Kecognition y.srem that identifies the utterance and trigger:; its associared dital code.

1 T VIliN Scil OlIlllIlilli LcltiU I,ySitIll c iT! tUilllS i U. i i i U i ill \\llit 11 another preferred meails for altering the ave pattern of the nouth microphone is to.,.c u a. c ", . ; t fi 0.,; thc I,L. v cv nk cI a c, ,,,chal1l I,, Il.c" moNres a recording head fonard and lackcrc! along a constant moNing magnetic meillm to recorcl. the sound fi o!n the. roice microphone at a fixed fimda!nenta! frequency on to effectiN ely create a constant Nroice w-aN:e pattern of any person that canbe fed to the ADC and the roice recoonitionDroramme inthe comouter.

1 A UniN'ersal C'ornmunications Systetn as in Claim 14 in which the recording mediutn is a magnetic tcordiTl; disk, and the recordin head is attached to an arm that pis ots around the driN,e of the constant speed maletic disk and which moN-es against the diTectio' ofthe disk spin to lowet- the frequency and in the direction of the spin to raise the frequency and the mociified info'ntion transfeTTeci to the disk is retrieved by a pickmp head and is then erased by an erase head which are UUill CiUS il, CcICIl TIU DiCli cU-t pOSiiiOllCU iTl Ct SUiiit 1Led pUSiiiOTI ON CT ill disl: <r A Uni Y, sal C'Gn,,r,unic.uiorls,ycr,, as iti 'ai,,, 1 ir, hich c, cco, dir,E medium is a continuous loop magnetic tape or wire running at a constant speed at!d the recorctino hearl rn move hack nCI fotth h means of a servo mechanism cotrolled by the PLL.

17] A Universal Ciommunications System as in (2laims 14.15 &lb in which the mono-tone voice sound as an analogue wavefortn that is retrieYed by the pickup head is fed to the ADC where it is satled digitally and fed to the Yoice recognition system. where it is p'cessed to ide'dify the equiYalent sentence, phrase or word' and its equialent code.

id}A UniYersai communications syrstem as in Ciaim i anci any ofthe precedin, claims in which only The Code or identification tag. is transmitted o'er any IIItUiUIII HuL'h c' TtlCilUIl,irC CvlC illC;Udill iiL' i'iUI, Ul-'V.Y'il-Cl' U satellite communication and since the system transmits a Digital C'ode only, it , A A-_ I 1 1 A r A: . I I "'t.a c' v., Y.;tCI&, CU.'l 1 ''11 spac ",- a''--vv;'c",,.,a.J multiplexed or may be transmitted as packets as used on the Internet, travelling !o" mny =!i!h!e outes to its destitltion.

19) A (Tnirersal Communieatioms System as in ens of the preceding Claims in wiuch all the participatinu persons hae a similar system and the common Code or identificarion of any utteraee c cu trigger their receisers to speak the same uttera'ce in each persols lanuae that has been preNiolisk recorded anti programmecl into its ss ster,l.

O)A lTni-ersal Communicatiors Svstem in an, ofthe prececling {?laims in vvhich in the eae of lectures Ot' new-s-ca.sting, the coc is opti otlai iy not ansmitteci ut is fiPr| into a pre-prof^minecl nnni-corr!ter to trier apre-reco.=ded equiralent lliiCl It ti ill i.ll oic U1;tIi iU1 t'ti Ui 1N Cl!Cll iOl-t'igll iOllUt.

_ l A tTni ersal Communicatiols Svstein as in ay of t he preN ious iinU whicli C'0.93;'iSCS l oabl; CG;,;j,ei Ov. S;ii.t'*lc' Gt h-Cil' .- tG '3; 'Ci - i CG;ii,ti.j;-"it house. a ni!n-cosurei pi-ograted vitt, the -o.e reco1itiot'xysten- thcat triggers the pre-prog-amuned cotle that is then stored into an independent System within the Box for use when it is required to utter sentences in a foreign language.

22) A Universal Communications System as in Claim 19 in which the equivalent Uii-tiIILt 1U i1Cil- CULTS ill Iy 101 Ci=1 iLIllUC -U C-1 CLOI it'll IL; SiLIltU 011 any suitable porthole memory system such as a floppy disk or CD-RVM or a hard disk etc. Id cm, be loomed into t.hc in 3pc..dent Systc... when.q;;i cd 'Loire the system to be used in many coveys such as when traN;ellin, to foreign countries or Oi,i!!ectlres Id tasks in a foreign tongue.

2:) A UniN;e sat Communications System as Claims 21 &22 in which She Digital Code is transmitted either by weak short distance radio signals or by a connecting wire to the mini-computer of the other party in face to itace conversations to triter the equivalent utterance in his own lal, uwge.

24} A (universal 1Jomninnications System as ill any of the previous Claims in which the processing programme can be less complicated, using less processing memory than in the normal dictation systems id use such that the processing tasks call be increased for the equivalent memory available in basic computers and the T-SvilS' iilTItS LIT gUiL'K41- Tlil 1-4 iilil4 Sp44T iTld411d4Ui iT-HTTHiiivTl reality.

A T TV; 7 A_ - A I At A; AA i; A. - LEA; A i A A _ A. ' I. ; . . Am as v''' 41 Am vtt''lllil4atv''la - L4.1. ..1.y V1 1..4 I,' 4 vl;o 11. ... VVII;II it is only necessary for a person speaking any foreign language to teach their 'oiee R erognition prom ample all the sentences phrases and chords in the common Lexicon together with their equivalent Code in order to communicate with and understand any foreign sneaker.

26is A lTniversal Communications System as in any of the previous Claims substantially as described in the Text and the Claims and with reference to the Diagrams and which in no way limits the applications described in the text.

Amendments to the claims have been filed as follows I) A universal communication system comprises in combination, Lexicons in a plurality of languages, and each Lexicon includes sentences, phrases and/or words in one language only and which have the same common meaning in all the Lexicons of the plurality of languages, an identification Marker associated only with the common meaning that is allocated and which acts as a link between all of the common meaning sentences phrases and/or words in all the Lexicons in all the plurality of languages, a Speech Recognition Means in each language that has been taught the common meaning utterances of its own language and which can recognise the speaker's utterance in his own language and means is provided for only the associated Marker to be transmitted to the receiver, Triggering Means to use the transmitted Marker to replay a pre-recorded utterance or text of the common meaning sentence, phrase and/or words in the receivers language, and Transmitting Means for sending the Marker to the receiver wherein the speech recognition means further comprises means to adjust the speakers voice wave form to a fixed fundamental frequency that stabilises the speakers voice pattern formants.

2) A universal communication system as in Claim 1, in which the Lexicons include normal everyday telephone and mobile chat and conversations that can be understood by a majority of the people, to more advanced conversations, broadcast speeches, lectures and specialized topics such as science and medicine and the like.

3) A universal communications system as in Claims 1 and 2, in which the common meaning is translated faithfully in any other language or dialect regardless of length or complexity of the word sequences to achieve a true translation and comprise discrete blocks common to the culture and language form that convey the essence of the conversation.

4) A universal communication system as in Claims 1 to 3, in which the contents of the Lexicons in each language wherever possible comprise the main stem of the sentence or phrase and the key words that change the meaning are then added with the proviso that the aggregate stem and key words form a coherent sentence or phrase that does not require any further mental translation input and each component can have its own Marker that can be sent in the proper sequence to comprise a complete sentence or phrase that can be understood in all languages.

5) A universal communication system as in Claims I to 4 in which all the Lexicons in every language are periodically updated at the same time with new material to meet the requirements of its users and eventually it takes on the semblance of a compromise language form with its own style and which takes into account the cultural differences and the like when selecting the common meanings and in which the aim is to achieve a goal in which all the discrete blocks selected in all the Lexicons represent a common meaning that can be transposed in any language 6) A universal communication system as in Claims I to 5 in which a predominant language such as English is selected as the lead language from which all the utterances in all the Lexicons are derived and which can be modified by feed back to incorporate the equivalent utterances in the other languages.

7) A universal communication system as in Claim 1, in which the Marker can be a code, number or other identifying label that is arbitrarily allocated and has no relationship with the utterance's voice pattern characteristics but is permanently associated only with the individual common meanings in all the Lexicons and represents a link between all the common meaning utterances in the plurality of 1+ languages and when the utterance is recognised only its allocated marker is transmitted in digital or analogue form over any communication system.

8) A universal communication system as in Claims I and 7, in which for example in English the Marker for "good morning" is arbitrarily allocated the number "l" and this number when transmitted to any user or between all users of the universal system will trigger the common meaning utterance in any language and for example a German receiver will hear "guten morgen" 9) A universal communication system as in Claims 1,7 and 8 in which the marker is encrypted by mutual consent between the parties for security reasons.

] O) A universal communications system as in Claim I in which in the Speech Recognition Means may be speaker dependent or speaker independent and does not require the correspondents to learn a foreign language or dialect but all the work required is to teach their speech recognition systems as required in their instructions or alternatively with all the sentences phrases, and/or words of their own language Lexicon and the systems have means to attach the relevant Markers and as soon as the speaker's utterance is identified, the translation part of the whole exercise is complete except that means is provided for the marker to be sent automatically to the receiver.

I 1) A universal communication system as in Claims I and 10, in which the Speech Recognition Means is not required to perform a mental act but is only required to match the Lexicon utterances using any commercial system that may include only matching of templates or other identifying parameters of the voice pattern of the discrete blocks of the utterances taught to the system in as much as the Lexicons in all languages have already translated the words, phrases and sentences solving the identification difficulties presented by domain dependence and ethnic speaker variability.

12) A universal communications system as in Claims I and I 1, in which the same speaker variability is eliminated by replacing the single microphone with two microphones one of which is positioned in the region of the Larynx (Adam's apple) to pick up the fundamental frequency waveform and the other is positioned near the mouth to pick up the complex spoken waveform that includes the fundamental frequency 13) A universal communication system as in Claims I and 12 in which the sound from the Larynx that comprises the fundamental frequency that changes up and down as we speak normally and is in he form of a simple sine wave, is constantly monitored, and any deviation of this frequency from a fixed pre-determined frequency is determined continuously.

14) A universal communication system as in claims 1, 12 and 13 in which the deviation from the fixed frequency produces a feed-back current that is used to continuously alter the ever changing fundamental frequency of the complex wave pattern (Formants) of the spoken waveform that include the harmonics back to the level of the fixed fundamental frequency, using any suitable feed-back techniques such as a Phase Locked Loop Detector (PLL) used in radios and the adjusted wave form of the utterance is fed to the Analogue to Digital Converter (ADC) of the Speech Recognition Means 15) A universal communication system as in Claims 1, 12 to 14, in which means is provided to cause the feed-back current to vary the sampling speed of the ADC, effectively causing the complex waveform to assume the monotone profile.

16) A universal communication system as in Claims], 12 to 15 in which mechanical means is provided which includes a constant spinning magnetic disk, or a continuous loop magnetic tape or wire that has recording means which in the case of the disk, is

IS

fixed to an arm that rotates around the spindle of the disk and is capable of rotating forward or back under the control of the feed-back from the PLL to record the adjusted voice waveform, followed by a fixed pickup means to extract the monotone utterance that is passed on to the ADC, and then immediately by a fixed erase means to clear the recording from the disk for re-use 17) A universal communication system as in Claims 1, 10 and 1 1 in which there is provided means to over-ride the recognition system and insert utterances transmitted by normal communication means that are not in the Lexicons such as names of people, specialist words and universal words such as "coca cola" and the like 18) A universal communication system as in Claim 1, 10 and I I in which there are less parameters required for recognising the utterance and simple independent speaker recognition programs using less memory are now a reality and the speech recognition computing means can therefore be portable or transferred to a special ISP or third party computing service and the like 19) A universal communication system as in Claim 1, in which the Triggering Means pre-records into any suitable audio storage system all the utterances of the words, phrases and sentences in its own language Lexicon in a retrievable form together with their associated Markers and has means to use the Marker that can be transmitted to the receiver by the sender that is the same as the associated Marker such that when the Markers match each other it triggers the replaying of the pre-recorded utterance of its equivalent word, phrase or sentence back to the receiver in his own language.

20) A universal communication system as in Claims I and 19 in which the receiver has the option when getting additions to the Lexicon for teaching his speech recognition system to also use it to manually record his own voice together with the Marker into the audio storage system.

21) A universal communication system as in Claims 1, 19 and 20 in which the marker transmitted to the receiver can trigger its equivalent marker that has been downloaded into the triggering means that causes the writing of a pre-programmed text in his own language of the common meaning utterances on to a typewriter, a computer screen or any type of telephone with video or text link and the like.

22) A universal communication system as in Claim 1, in which the Transmitting Means comprises any communications systems that use any medium efficiently and includes telephone wire or cable or optical fibre or wireless or satellite communication in the form of digital or analogue form, directly or in packets over the Internet including multiplexing and 3G and WAP technology and the like and the case of a video link or TV broadcast, the utterance latency period between speaker to receiver can be synchronised by delaying the picture to make it more realistic 23) A universal communication system as in claims 1 and 22 in which the Transmitting Means comprise hardware that includes any type of computer such as desktop, laptop or mini computers, fixed or mobile telephones, broadcast wireless and television or any other person to person electronic equipment that can use the universal communication system 24) A universal communication system as in any of the preceding Claims in which all the participating persons worldwide have a similar type of system and the common meaning Marker associated with any utterance in any of the common Lexicons can trigger the equivalent utterance in any other language.

25) A universal communication system as in any of the preceding Claims when used for lectures, conferences or large meetings and the like, the Marker optionally is not transmitted but is sent to a receiving system nearby to trigger a pre-recorded voice of l6 an ideal speaker of any chosen language and if the audience is multilingual, individual receivers programmed in any language can be provided for each recipient.

26) A universal communication system as in any of the preceding Claims in which any additions to the Lexicon including their Markers are centrally or regionally prepared and are sent to all users by any means preferably in text form so that users may teach their own Speech Recognition Means and record into their Triggering Means their own voice and optionally the utterance additions may be downloaded in the form of a pre-recorded voice of an ideal speaker together with the equivalent text including their common meaning Marker that has also been also centrally prepared direct via the interact or other medium.

27) A universal communication system as in any of the preceding Claims that comprises a portable version of the system such as a mini-computer that has the Triggering Means which can be pre-loaded from any portable storage system which may include a CD or floppy disk with the Lexicon together with the Marker in any foreign language for speaking only such as for tourists or otherwise and if the receiving person has a similar system in his own language equipped with electrical wire contact or a short distance wireless transmitter between sender and receiver they may converse directly with each other.

28) A universal communication system as in any of the preceding Claims for use in cases where there is no available physical communications system except wireless contact and the like in which the receiving persons has only a receiver such as a radio / Triggering Means combination programmed by any suitable means with their own language Lexicon and Marker, and they may listen to any news or other communication transmitted in any language or means is provided to allow any other use such as for receiving road directions and warnings and the like in towns and the like 29) A universal communication system as in any of the previous Claims substantially as described in the Text and Claims with reference to the Diagrams and which in no way limits the applications herewith described.