[go: up one dir, main page]

WO2011016042A1 - Reconnaissance et identification de tonalités de retour d'appel - Google Patents

Reconnaissance et identification de tonalités de retour d'appel Download PDF

Info

Publication number
WO2011016042A1
WO2011016042A1 PCT/IL2010/000637 IL2010000637W WO2011016042A1 WO 2011016042 A1 WO2011016042 A1 WO 2011016042A1 IL 2010000637 W IL2010000637 W IL 2010000637W WO 2011016042 A1 WO2011016042 A1 WO 2011016042A1
Authority
WO
WIPO (PCT)
Prior art keywords
database
audio
call
sampling
clips
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.)
Ceased
Application number
PCT/IL2010/000637
Other languages
English (en)
Inventor
Shaul Wilenski
Avey Gabrielli
Avraham Kadosh
Eyal Katz
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.)
He Catalyst Ltd
Original Assignee
He Catalyst Ltd
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 He Catalyst Ltd filed Critical He Catalyst Ltd
Publication of WO2011016042A1 publication Critical patent/WO2011016042A1/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M3/00—Automatic or semi-automatic exchanges
    • H04M3/42—Systems providing special services or facilities to subscribers
    • H04M3/42017—Customized ring-back tones

Definitions

  • the present invention relates to apparatus and a method for music matching within telephone networks and, more particularly, but not exclusively to identification of ring back tones.
  • ring back tones or CRBT
  • the tones play to callers instead of the standard tones before the call is answered. If a caller likes the ring back tone he is listening to, then as long as he and the caller are in the same network, he is often able to order the tone for himself, for example by pressing a particular DTMF key on his phone.
  • the known system uses the network operator's database of users and ring back tones in order to determine what ring back tone is being ordered.
  • the network operator's database cannot be used since the callee 's CRBT is not known to the caller's operator.
  • databases are not always available, or may be distributed in such a way that there is no central information store or the databases may need to be compiled or updated.
  • the present embodiments relate to a method and apparatus for identifying ring back tones automatically from a callee at a first telephone network so as to make them available to a caller at the same telephone network or to allow the network to compile a database of what ring back tones it has available.
  • apparatus for audio matching within a telephone network comprising:
  • sampling unit located at said telephone network for sampling audio being played in response to a call involving said telephone network, the sampling unit being adapted for sampling of signals within a telephone network frequency range;
  • an analysis and matching unit for matching the sample with a plurality of audio clips held on an audio database to find a matched clip
  • Video clips may be identified via the audio track, but may be identified via a fingerprint of the video as well. Using the former, without video identification would not distinguish between different video clips having the same music, but the latter would allow for differentiation between say the original clip and a spoof version.
  • An embodiment may comprise a call placement delay unit for delaying placement of a call until an announcement has been made to a caller.
  • the call placement delay unit is configured to operate for call destinations previously recorded as having call ring back tones (CRBT) that match the tunes on the tunes database.
  • CBT call ring back tones
  • An embodiment may comprise an IWR (Interactive Voice and Video response) system including a DTMF detector and Speech recognition resources and capability for detecting a caller response to a CRBT currently being played, the assignment unit being operatively connected to the DTMF detector to assign the matched tune according to the caller response.
  • IWR Interactive Voice and Video response
  • An embodiment may comprise a callee database indicating out-of-network callees known to have an identified CRBT.
  • sampled tunes are categorized into tunes for which distribution rights are owned, tunes for which distribution rights are available but not currently owned and tunes for which distribution rights are not available, and the callee database is updated with out-of-network callees, corresponding sampled tunes and corresponding categories.
  • An embodiment may comprise a call placement delay unit for delaying placement of a call until an announcement has been made to a caller, and wherein the call placement delay unit is configured to operate only where the callee appears in the callee database and wherein the corresponding category indicates a tune for which distribution rights are owned.
  • An embodiment may include a silent mode of operation, wherein the callee database is updated but the assignment unit is not operated.
  • the tunes database comprises tunes for which distribution rights are owned, tunes for which distribution rights are available but not currently owned and tunes for which distribution-rights are not available.
  • An embodiment may be configured to refer requests relating to tunes not present on the database to an operator, thereby to facilitate manual updating of the tunes database with popular tunes not currently being held.
  • An embodiment may be configured with a rights acquisition unit for gathering statistics regarding requests relating to the tunes for which distribution rights are available but not currently owned, and, when the requests exceed a predetermined threshold, seeking to obtain the distribution rights.
  • the matching unit is configured to obtain a fingerprint of the sample and compare it to fingerprints of the tunes on the tunes database, to find closest candidate matches, and then to carry out detailed searches of the closest candidates to obtain the match.
  • the DTMF detector and speech recognition resources and capability are in operation for all calls, irrespective of whether the call was subjected to call placement delay and an announcement.
  • the music being sampled is part of a video clip, and whereby the matching is carried on the tune to match with the video clip.
  • the sampling unit is configured to sample the music at a sampling rate of substantially 8KHz.
  • a method for audio matching including music matching, to provide a caller with a music sample heard as a call ring back tone from a callee at a second network, the method comprising: at the first telephone network sampling music being played as a call ring back tone in response to a call being placed, the sampling being adapted for sampling of signals within a telephone network frequency range;
  • the method may be used to establish what audio material is available on the network.
  • Implementation of the method and/or system of embodiments of the invention can involve performing or completing selected tasks manually, automatically, or a combination thereof. This refers in particular to tasks involving the control of telephone server equipment.
  • a data processor such as a computing platform for executing a plurality of instructions.
  • the data processor includes a volatile memory for storing instructions and/or data and/or a non-volatile storage, for example, a magnetic hard-disk and/or removable media, for storing instructions and/or data.
  • a network connection is provided as well.
  • a display and/or a user input device such as a keyboard or mouse are optionally provided as well.
  • FIG. 1 is a simplified block diagram illustrating two networks and a music matching unit for matching out of network CRBTs according to a first embodiment of the present invention
  • FIG. 2 shows in greater detail the music matching device and CRBT database of the device of FIG. 1;
  • FIG. 3 is a simplified flow chart illustrating the operation of the call delay unit of FIG. 2;
  • FIG. 4 is a simplified flow chart showing how a CRBT is assigned to the caller following sampling according to an embodiment of the present invention
  • FIG. 5 is a simplified diagram showing a variation of the embodiment of FIG. 1 in which decentralized databases are available either instead of or in addition to a centralized database;
  • FIG. 6 is a simplified diagram in which processing for CRBT recognition is split between a user client and a remote location, according to an embodiment of the present invention.
  • the present embodiments comprise a device and method for music matching across telephone networks and, more particularly, but not exclusively to identification of ring back tones.
  • music matching may refer also to audio matching in general and video clip matching, as will be explained below.
  • FIG. 1 illustrates apparatus for music matching in a telephone network 8.
  • a caller 10 calls a callee 12.
  • Network 8 has an isolated database 14 of call ring back tones CRBT to which callee 12 belongs, and Network 8 has a centralized database 16 of call ring back tones which does not know about the isolated database 14.
  • Caller 10 calls callee 12 he hears a call ringback tone that callee 12 has selected from his isolated CRBT database 14.
  • Caller 10's network B has no access to CRBT database 14 so caller 10 has no way of selecting the ringback tone that he hears, should he like the tune and wish to use it for himself.
  • network 8 has no way of adding the tunes to its centralized database 16 should it so desire.
  • a music matching unit 18 is provided.
  • the matching unit samples the ring back tone and attempts to find a match amongst available CRBTs at central CRBT database 16. If a match is found then the matching CRBT can be noted at central CRBT database 16 and/or may be assigned to caller 10 as his operational CRBT.
  • One idea is so that telephone network 8 may keep track of CRBTs on the network. Additionally or alternatively, an idea is to encourage impulse purchasing by subscribers. A potential customer hears the product and is able to react immediately with a purchase.
  • the music matching unit 18 comprises a sampling unit 20 located in association with network 8, which listens to incoming traffic in order to sample music played as a ring-back tone. Ring back tones are played instead of a standard tone while waiting for the callee to answer.
  • the sampling unit 20 preferably signals at the Nyquist rate for the telephone network. Telephone networks operate with a baseband of up to 4KHz so the Nyquist rate is 8KHz. In this sense the sampling unit is adapted for the telephone network frequency range.
  • An analysis and matching unit 22 is connected downstream of the sampling unit 20 and matches between the sample and tunes held on CRBT database 24 to find the best match.
  • a fingerprint is extracted of the sample and is compared with preformed fingerprints of the tunes in the database for fast matching to find a subset of good candidates. Then more detailed searching is done to find a closest match from the good candidates. If no match is close then the matching returns a fail.
  • an assignment unit 26 may assign the matched tunes as playback tones to the respective subscribers. As will be discussed in greater detail below, such an assignment may not always be possible, since usage rights for the tune may not always be owned. In the latter case the assignment unit may take note of popular requests for which rights are not available so that rights may be purchased.
  • Call placement delay unit 28 listens to outgoing calls within the network and is able to delay placement of a call until an announcement has been made to a caller.
  • the announcement may briefly tell the caller about the service, and indicate which key to press in order to obtain the ring-back tone he hears for himself. As soon as the announcement is made the call is put through and the callee's ringback tone is heard.
  • call placement delay unit 28 need not interfere with the operation of sampling unit 20.
  • the service provided by the present embodiments is intended to work in the background at all times, whether announced or not. More knowledgeable users who remember the key press, can use the service even when not announced.
  • the keypress since the keypress is always available during the ringing phase of a call it may use a key combination that can be accepted unambiguously as indicating an intention to obtain the CRBT.
  • the double keypress "**" was used in a prototype system.
  • the call placement delay unit 28 is connected to callee database 30.
  • Callee database 30 contains telephone numbers that have previously been called, and records whether the numbers have ring back tones at all. If the callee numbers have ring back tones, then the callee database indicates those who have tones that are recognized on the current network, and particularly those who have tones that can be provided by the current network.
  • the call delay unit necessarily makes its announcement before any call has been put through and hence before any CRBT has been played. An announcement would be superfluous and cause user irritation if made where there is no following ring back tone.
  • the caller's network may have no direct way of knowing whether callees have tones currently set in the absence of access to all the isolated databases.
  • the problem may be dealt with by storing numbers and call back tone status in the callee database.
  • the call placement delay unit restricts announcements to those numbers on other networks previously recorded as having call ring back tones (CRBT) that match tunes on the tunes database and available in the current network.
  • CRBT call ring back tones
  • the tendency of users who have CRBT tones to retain them leads to an inertia on the network which means that most of the time the announcements will not be superfluous.
  • a DTMF detector 32 detects a caller's response to a CRBT currently being played. The DTMF detector directs the assignment unit 26 to provide the tune to the subscriber or to indicate to the system that distribution rights should be purchased or to carry out any other appropriate response.
  • the callee database may thus indicate callees known at least recently to have an identified CRBT.
  • Sampled tones comprising tunes or music in general, are categorized for the purposes of the callee database into tunes for which distribution rights are owned, tunes for which distribution rights are available but not currently owned and tunes for which distribution rights are not available.
  • Non-availability may be due to the fact that the rights owner has signed an exclusive distribution agreement with someone else, or due to the fact that the minimum price available is too high to be. justified by the charge made to users on the network, or for a variety of other reasons.
  • the callee database is updated with callees, corresponding sampled tunes and corresponding categories for the tunes so that the announcements can be placed most effectively and cause minimum annoyance to the network subscribers.
  • the call placement delay unit may also store and use data of the caller, say for personalization of the service. Thus announcements may be rationed to one or two a month to any given caller, but with more announcements made to users who have shown interest in the service or have CRBTs themselves, and fewer announcements to users who ignore the service over time and evidently have no interest in CRBTs.
  • the call placement delay unit 28 may thus delay placement of a call to a callee until an announcement has been made to a caller.
  • the call placement delay unit 28 may operate specifically in the case that the callee appears in the callee database along with a data entry indicating a CRBT for which distribution rights are owned.
  • the data entry may at best reflect the status of the callee at the previous occasion on which that callee was called from within the network
  • the callee may have changed or cancelled his CRBT by the time the present call is placed.
  • the general tendency to inertia on a network ensures that the announcement will be effective most of the time.
  • the system may additionally include a silent mode of operation in which no announcements are made.
  • a silent mode of operation sampling and matching of tones is carried out in the normal way but the results are merely used to fill in data in the callee database.
  • Using such a silent mode in the initial stage enables a large number of well-directed announcements in a following, active stage.
  • the assignment unit can either be switched off altogether or can be kept operating in the background, allowing knowledgeable users to select
  • the silent mode may be used for initial set up of the system, but also may be entered subsequently to allow updating of the database if it is found that unacceptable numbers of inappropriate announcements are being made.
  • the CRBT database 24 may include not just tunes for which distribution rights are owned, but also tunes for which distribution rights are available but not currently owned and tunes for which distribution rights are not available. Thus the system is able to match tunes which are not available and thus identify demand, for which rights can then be purchased. In this way the system allows for rapid updating say to include newly popular songs.
  • the analysis and matching unit 22 may include a feature for referring requests relating to tunes not present on the database to an operator. The operator may then attempt to manually recognize the tune and take action as necessary to add the CRBT to the database. Typically the operator concentrates on repeated requests so as to update the database with popular tunes.
  • the ability to obtain popular tunes and associated rights may be assisted by a rights acquisition unit 34 which monitors unfulfilled requests, and gathers statistics. For example if requests relating to a given CRBT for which distribution rights are available but not currently owned exceed a threshold number or rate, then a trigger is set to seek to obtain the relevant distribution rights.
  • Videophones which supports video calls, may play video clips in place of CRBTs and the present embodiments allow a music video clip to be identified by extracting the sound track and processing the sound track in exactly the same way as described above with a CRBT.
  • the corresponding video clip is thus identified through its sound track.
  • a fingerprint of the video may be used as well or instead. Using the former, without video identification, would not distinguish between different video clips having the same music, but the latter would allow for differentiation between say the original clip and a spoof version.
  • Fig. 3 is a simplified flow chart illustrating the procedure when a caller places a call within a network using the embodiments of the present invention.
  • a caller calls a callee. If the callee is listed in the database as having an available CRBT then decision box 42 returns a yes and decision box 44 is entered. Decision box 44 checks that the present caller is due an announcement. This may be because he has never been given an announcement before or because he has not been given an announcement this month. If the caller is due an announcement then the call connection is delayed in stage 46 and an announcement about the service is played. In stage 48 the call is connected, irrespective of whether the yes or no paths were taken from the decision boxes.
  • Fig. 4 shows the procedure when the CRBT is sampled. Sampling takes place in stage 50. If the sample is identified then decision box 54 checks whether rights are owned. If the rights are owned then the identified
  • CRBT is assigned in box 56 to the caller so that it is played next time someone calls him or her.
  • box 54 if the rights are not owned then flow proceeds to the rights acquisition procedure beginning with decision box 62 which sets a popularity test for the particular tune. There is little point trying to obtain rights for a tune if only one user wants the rights. On the other hand if there have been tens or hundreds of requests in the past hour then it is high time that the current network has this tune on its availability list.
  • a threshold is set, which may be an absolute number or may be a rate of requests, and if the threshold is reached then the system attempts to purchase rights in box 64.
  • the user request is set to a pending state in box 66 and the tune is assigned to the user in box 56 as soon as the rights are obtained. If the threshold is not met then the request is refused on box 68.
  • Fig. 5 is a simplified diagram showing a variation of the embodiment of Fig. 1 in which a decentralized database 70 is available to the music matching unit 18.
  • the decentralized database may be local or may be based at any available location and is particularly suitable for VoIP based telephone networks and SIP clients, which do not tend to have centralized resources.
  • the database 70 may be one of a series of non-centralized databases available to the music matching unit, and the databases may belong to different and even competing entities.
  • the music matching unit may for example provide for online purchase of the recognized CRBT or the song title, including single click purchase.
  • Fig. 6 is a simplified diagram showing an embodiment of the present invention for use in a distributed system, particularly suitable for VoIP and other systems where the user terminal has intelligence.
  • the incoming CRBT is sampled by sampler 80, and features are extracted by feature extraction unit 82 at a user client 84.
  • Reference numeral 86 denotes the remote location at which the music matching unit 18, or centralized or decentralized databases 16 and 70, are located.
  • Feature ID unit 88 at remote location 86 tests the extracted features against the CRBTs in the database, to find a match.
  • CRBT commands unit 90 then allows the match to be provided to the client or whatever other function may be required regarding the CRBTs.
  • the billing and management unit 92 carries out housekeeping functions including billing.

Landscapes

  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Telephonic Communication Services (AREA)

Abstract

La présente invention concerne un appareil servant à établir des concordances musicales à l'intérieur de réseaux téléphoniques. Cet appareil comprend un module d'étalonnage intégré au réseau téléphonique de façon à étalonner un morceau de musique en train d'être joué en tant que tonalité de retour d'appel en réaction à un appel, le module d'étalonnage étant conçu pour étalonner des signaux à l'intérieur d'une plage de fréquences du réseau téléphonique. Cet appareil comprend également un module d'analyse et d'établissement de concordances servant à établir des concordances entre l'échantillon et une pluralité de tonalités conservées dans une base de données de tonalités de façon à trouver une concordance, à la suite de quoi la concordance peut être utilisée pour constituer des données permettant de savoir quels sont les tonalités disponibles sur le réseau. En outre, un module d'affectation peut affecter à des abonnés intéressés des tonalités pour lesquelles des concordances ont été trouvées.
PCT/IL2010/000637 2009-08-05 2010-08-05 Reconnaissance et identification de tonalités de retour d'appel Ceased WO2011016042A1 (fr)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
US23134709P 2009-08-05 2009-08-05
US61/231,347 2009-08-05
US26138409P 2009-11-16 2009-11-16
US61/261,384 2009-11-16

Publications (1)

Publication Number Publication Date
WO2011016042A1 true WO2011016042A1 (fr) 2011-02-10

Family

ID=43063778

Family Applications (2)

Application Number Title Priority Date Filing Date
PCT/IL2010/000636 Ceased WO2011016041A1 (fr) 2009-08-05 2010-08-05 Reconnaissance et identification de tonalités de retour d’appel
PCT/IL2010/000637 Ceased WO2011016042A1 (fr) 2009-08-05 2010-08-05 Reconnaissance et identification de tonalités de retour d'appel

Family Applications Before (1)

Application Number Title Priority Date Filing Date
PCT/IL2010/000636 Ceased WO2011016041A1 (fr) 2009-08-05 2010-08-05 Reconnaissance et identification de tonalités de retour d’appel

Country Status (1)

Country Link
WO (2) WO2011016041A1 (fr)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116259292B (zh) * 2023-03-23 2023-10-20 广州资云科技有限公司 基调和音阶的识别方法、装置、计算机设备和存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2002060162A2 (fr) * 2000-11-30 2002-08-01 Enterprise Integration Group, Inc. Procede et systeme permettant de prevenir l'amplification d'erreur dans des dialogues en langage naturel
WO2006057504A1 (fr) * 2004-11-25 2006-06-01 Widerthan Co., Ltd. Systeme et procede de copie de la tonalite de transfert d'un signal de retour d'appel
WO2006117777A2 (fr) 2005-04-29 2006-11-09 Hingi Ltd. Procede et appareil de fourniture de donnees de contenu
US20080167921A1 (en) * 2007-01-09 2008-07-10 Nhn Corporation Advertisement method using ring tone providing service and advertisement system thereof, and computer- readable media for recording advertisement method program

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2002060162A2 (fr) * 2000-11-30 2002-08-01 Enterprise Integration Group, Inc. Procede et systeme permettant de prevenir l'amplification d'erreur dans des dialogues en langage naturel
WO2006057504A1 (fr) * 2004-11-25 2006-06-01 Widerthan Co., Ltd. Systeme et procede de copie de la tonalite de transfert d'un signal de retour d'appel
WO2006117777A2 (fr) 2005-04-29 2006-11-09 Hingi Ltd. Procede et appareil de fourniture de donnees de contenu
US20080167921A1 (en) * 2007-01-09 2008-07-10 Nhn Corporation Advertisement method using ring tone providing service and advertisement system thereof, and computer- readable media for recording advertisement method program

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
411-SONG: "411-SONG: How It Works?", INTERNET CITATION, 22 March 2008 (2008-03-22), pages 1 - 7, XP007915783, Retrieved from the Internet <URL:http://web.archive.org/web/20080322034224/www.411song.com/HowItWorks.asp> [retrieved on 20101116] *
WANG, AVERY: "THE SHAZAM MUSIC RECOGNITION SERVICE", COMMUNICATIONS OF THE ACM, vol. 49, no. 8, 31 August 2006 (2006-08-31) - 31 August 2006 (2006-08-31), pages 44 - 48, XP002610302, Retrieved from the Internet <URL:http://www.doc.ic.ac.uk/~ih/teaching/group-projects/proposals/dfg/p44-wang.pdf> [retrieved on 20101119] *

Also Published As

Publication number Publication date
WO2011016041A1 (fr) 2011-02-10

Similar Documents

Publication Publication Date Title
KR100866859B1 (ko) 인터넷 전화 음성 메일 관리방법
US10110741B1 (en) Determining and denying call completion based on detection of robocall or telemarketing call
US9313307B2 (en) System and method for verifying the identity of a user by voiceprint analysis
US7986776B2 (en) Method and system for telephone wait user interface selection
US20070258577A1 (en) Method and system for telephone wait user interface selection
US11232461B2 (en) System and method for causing messages to be delivered to users of a distributed voice application execution system
CA2352981A1 (fr) Methode de modification de la parole permettant a l&#39;utilisateur de selectionner un dialecte
US20080008306A1 (en) Out of band messaging for completion of response to information request
US20060188087A1 (en) System and method for caller-controlled music on-hold
US11641420B2 (en) System and method for placing telephone calls using a distributed voice application execution system architecture
US8964960B2 (en) System and method for interacting with a user via a variable volume and variable tone audio prompt
US11909901B2 (en) System and method for displaying the history of a user&#39;s interaction with a voice application
US20100158218A1 (en) System and method for providing interactive services
US9210264B2 (en) System and method for live voice and voicemail detection
WO2010029301A1 (fr) Couplage téléphonie-informatique dans un environnement de fournisseur de services
US20160261743A1 (en) System and method for call progress detection
CN104732982A (zh) 交互式语音应答业务中的语音识别方法及装置
WO2011016042A1 (fr) Reconnaissance et identification de tonalités de retour d&#39;appel
US20160330323A1 (en) System and method for performing certain actions based upon a dialed telephone number
US9253301B2 (en) System and method for announcing and routing incoming telephone calls using a distributed voice application execution system architecture
EP2961146A1 (fr) Procédé et système de réglage des paramètres de détection dans un appareil de détection de musique d&#39;attente

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 10757636

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC DATED 28.06.12

122 Ep: pct application non-entry in european phase

Ref document number: 10757636

Country of ref document: EP

Kind code of ref document: A1