FR2778782A1 - Multi user voice recognition system - Google Patents

Multi user voice recognition system Download PDF

Info

Publication number
FR2778782A1
FR2778782A1 FR9806214A FR9806214A FR2778782A1 FR 2778782 A1 FR2778782 A1 FR 2778782A1 FR 9806214 A FR9806214 A FR 9806214A FR 9806214 A FR9806214 A FR 9806214A FR 2778782 A1 FR2778782 A1 FR 2778782A1
Authority
FR
France
Prior art keywords
voice recognition
recognition system
user voice
voice
multi user
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
FR9806214A
Other languages
French (fr)
Other versions
FR2778782B1 (en
Inventor
Henri Benmussa
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Individual
Original Assignee
Individual
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Individual filed Critical Individual
Priority to FR9806214A priority Critical patent/FR2778782B1/en
Publication of FR2778782A1 publication Critical patent/FR2778782A1/en
Application granted granted Critical
Publication of FR2778782B1 publication Critical patent/FR2778782B1/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/26—Speech to text systems
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/06—Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice

Landscapes

  • Engineering & Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)

Abstract

The system uses a number of preprogrammed keywords to access the user interface. A computer recognizes a key word spoken be the user to allow him or her to access functions that are controlled by the keywords.

Description

-4 2778782-4 2778782

Pour offirir au grand public un moyen naturel de dialogue avec un ordinateur, il a été développé des techniques de reconnaissance vocale. Deux types de systèmes existent à ce jour: - les systèmes monolocuteurs dédiés à un individu qui doit éduquer sa machine par sa parole, 3 - les systèmes multilocuteurs qui, en principe, t es mots de n'importe quel locuteur. Les pemies systèmes sont opérationnels mais réservés à un usage personnel. les seconds sont coûteux et ne reconnaissent qu'un nombre limité de mots et cela avec une fiabilité encore insuffisante. L'invention, objet du présent brevet, consiste à offrir un système multilocuteurs en ramenant ses techniques complexes à celles, bien matrisées, de la reconnaissance monolocuteur. Cette invention est particulièrement applicable à toutes opérations de télé pament tells que le télé  To provide the general public with a natural means of dialogue with a computer, voice recognition techniques have been developed. Two types of systems exist today: - monolocutor systems dedicated to an individual who must educate his machine by his word, 3 - multilocutor systems which, in principle, are the words of any speaker. The system pemas are operational but reserved for personal use. the latter are costly and recognize only a limited number of words, and this with still insufficient reliability. The invention which is the subject of this patent consists in offering a multi-speaker system by reducing its complex techniques to those, well mastered, of single-speaker recognition. This invention is particularly applicable to all tele pament operations such as tele

achat dans un grand magasin.purchase in a department store.

- Une personne désirant accéder au service vocal dn magasin, reçoit un code personnel et 1 doit lire un court texte standard permettant à ordinateur du magasin de constituer le  - A person wishing to access the voice service in the store, receives a personal code and 1 must read a short standard text allowing the store's computer to constitute the

dictionnaire vocal de ce nouveau client.  voice dictionary of this new client.

- lorqe le client veut effectuer, i distance, un achat, il compose noraement le code téléphonique du magasin, puis son code personnel. La machine réceptrice, connaissant l'identité du client, recherche dans son fichier le dictionnaire vocal de ce client, et peut anis: - confirmer, grâce à remreinte vocale, ldentité de son interlocuteur - utiliser le dictionnaire vocal pour interpréter, dune manière fiable, la voix de son interlocuteur.  - When the customer wants to make a purchase, at a distance, he normally dials the store telephone code, then his personal code. The receiving machine, knowing the identity of the customer, searches its file for the voice dictionary of this customer, and can anise: - confirm, by voice print, the identity of his contact - use the voice dictionary to interpret, in a reliable manner, the voice of his interlocutor.

L'invention consiste ainsi à créer un système mulmonolocuteur.  The invention thus consists in creating a mulmonolocutor system.

A titre d'exemple, pour réaliser un système destiné à un magasin de taille moyenne ( 10.000 clients inscrits au service vocal et faisant, en moyenne, un appel de 2 minutes). ln y aura, environ, une vingtaine d'appels sin tanés Un mode de réalisation consiste ià utiliser une carte CT ( voir Brevet certificat d'enregistrement N 98/24 NL du 12 Juin 1998) munie d'un microprocesseur TMS 320 C 6201, 1600 Mips, capable de gérer 50 reconnaissances vocales simultanée. Cette carte CT est disposée dans un PC. nuni de disques de stockage. Conmme un dictionnaire vocal occupe, environ, 0'1 Mo, la capacité de stockage doit donc tre de I Go, c'est i dire, une fraction de la capacité des plus petits disques du marché des PC. Un PC  For example, to make a system for a medium-sized store (10,000 customers registered for voice service and making, on average, a call of 2 minutes). There will be approximately twenty sin tan calls. One embodiment consists in using a CT card (see Patent registration certificate N 98/24 NL of June 12, 1998) fitted with a TMS 320 C 6201 microprocessor, 1600 Mips, capable of managing 50 simultaneous voice recognitions. This CT card is placed in a PC. provided with storage disks. Like an occupied voice dictionary, approximately 0'1 MB, the storage capacity must therefore be I GB, that is to say, a fraction of the capacity of the smallest disks on the PC market. PC

standard contenant 4 cartes CT pourrait desservir une clientèle d'ienviron 100.000 abonnés.  standard containing 4 CT cards could serve a clientele of around 100,000 subscribers.

Les applications industrielles d'un tel produit sont très vastes: télé achat, télé banque, télé action, commande vocale pour automobiles. Le fiable cot de tells machines permettra  The industrial applications of such a product are very vast: teleshopping, tele banking, tele action, voice control for cars. The reliable cost of such machines will allow

l'extension de ces services aux PMI.  the extension of these services to SMIs.

2 27787822 2778782

Claims (2)

Revendication sClaim s 1 Dispositif permettant i un grand nombre de locuteurs d'avoir leurs voix reconnues et  1 Device enabling a large number of speakers to have their voices recognized and interprétées par une machine commune.  interpreted by a common machine. CARACrERISE PAR l'asociation d'un dictionnaire vocal préenregistré, spécifique à chaque utilnateur et d'une  CHARACTERIZED BY the association of a prerecorded voice dictionary, specific to each user and a maclie générale de traitement.general treatment maclie. 2 Procédé de mise en oeuvre du dispositifselon la revendication 1, caractérisé en ce que la machine, avertie de l'identité de rappelant, accède ià son dictionnaire vocal et utilise, ainsi, les  2 A method of implementing the device according to claim 1, characterized in that the machine, informed of the caller's identity, accesses its voice dictionary and thus uses the programmes de la reconnaissance monolocuteur.  single speaker recognition programs.
FR9806214A 1998-05-18 1998-05-18 "MULTI-SINGLE SPEAKER" VOICE RECOGNITION SYSTEM Expired - Fee Related FR2778782B1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
FR9806214A FR2778782B1 (en) 1998-05-18 1998-05-18 "MULTI-SINGLE SPEAKER" VOICE RECOGNITION SYSTEM

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
FR9806214A FR2778782B1 (en) 1998-05-18 1998-05-18 "MULTI-SINGLE SPEAKER" VOICE RECOGNITION SYSTEM

Publications (2)

Publication Number Publication Date
FR2778782A1 true FR2778782A1 (en) 1999-11-19
FR2778782B1 FR2778782B1 (en) 2000-07-21

Family

ID=9526437

Family Applications (1)

Application Number Title Priority Date Filing Date
FR9806214A Expired - Fee Related FR2778782B1 (en) 1998-05-18 1998-05-18 "MULTI-SINGLE SPEAKER" VOICE RECOGNITION SYSTEM

Country Status (1)

Country Link
FR (1) FR2778782B1 (en)

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO1992000586A1 (en) * 1990-07-02 1992-01-09 Motorola, Inc. Keyword-based speaker selection

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO1992000586A1 (en) * 1990-07-02 1992-01-09 Motorola, Inc. Keyword-based speaker selection

Also Published As

Publication number Publication date
FR2778782B1 (en) 2000-07-21

Similar Documents

Publication Publication Date Title
US10019713B1 (en) Apparatus and method for verifying transactions using voice print
US8768709B1 (en) Apparatus and method for verifying transactions using voice print
US7515695B1 (en) Client customizable interactive voice response system
US6539084B1 (en) Intercom system
US20110282661A1 (en) Method for speaker source classification
GB2341705A (en) System and method for automatically providing financial services to a user using speech signals
US8761372B1 (en) Telephone channel personalization
CN108924218B (en) Method and device for pushing information
CN112181127A (en) Method and device for man-machine interaction
CA2500267C (en) Prepaid phone card service using speech recognition
US20050055310A1 (en) Method and system for accessing information within a database
US20140302814A1 (en) Centralized caller profile and payment system and methods for processing telephone payments
CN110581927A (en) Call content processing and prompting method and device
CA2231585A1 (en) Method of receiving a processing request by speech
US6700965B1 (en) Identifier-triggered personalized customer relations management service
FR2778782A1 (en) Multi user voice recognition system
JP2003519811A (en) Voice-activated identification of a user of a telecommunications line in a telecommunications network during a dialogue with a voice-activated dialogue system
EP1098493B1 (en) Method and suitable server for teleservices
CN116049411A (en) Information matching method, device, equipment and readable storage medium
FR2795266A1 (en) METHOD AND SYSTEM FOR SAFE AND FAST VOICE IDENTIFICATION OF A NOMED OBJECT EMITTING AN ACOUSTIC SIGNAL
US6125389A (en) Apparatus and method for providing network subscriber services
CN115442484A (en) Telephone traffic seat allocation method and device, computer storage medium and electronic equipment
US20070061395A1 (en) Gateway apparatus for information processing using mobile communication
CN111967868B (en) Micro-bank transfer method and device, storage medium and electronic equipment
CN115116467A (en) Audio marking method and device and electronic equipment

Legal Events

Date Code Title Description
ST Notification of lapse