PROCESSING

TECHNICAL FIELD

[0001 ] The present disclosure relates to call processing, and more particularly to a computer-implemented method for automated call processing, a computing device , and a computer program product .

BACKGROUND

[0002] Automated call processing can util ise various technologies , such as machine learning and automatic speech recognition, to improve the efficiency of call processing . However, there can be various situations in which a user calling a service utilising automated call processing is in a less than ideal environment or situation to interact with the service . This can make the automated call processing to not function properly and reduce the efficiency of the call processing .

SUMMARY

[0003] This summary is provided to introduce a selection of concepts in a s implif ied form that are further described below in the detailed description . This summary is not intended to identify key features or essential features of the claimed subj ect matter, nor i s it intended to be used to limit the scope of the claimed subj ect matter . [0004] It is an obj ective to provide a computer-implemented method for automated call processing, a computing device , and a computer program product . The foregoing and other obj ectives are achieved by the features of the independent claims . Further implementation forms are apparent from the dependent claims , the description and the figures .

[0005] According to a first aspect, a computer-implemented method for automated call processing comprises : receiving a call from a user ; identifying an environment of the user during the cal l using acoustic scene clas sification ; configuring at least one property of an automated call processing system according to the identified environment ; and processing the call at least partially using the automated call processing system . The method can, for example , improve the functionality of the automated call processing system in various different call conditions .

[0006] In an implementation form of the first aspect , the automated call processing system comprises an automatic speech recognition system and the configuring the at least one property of the automated cal l process ing system comprises configuring the automatic speech recognition system according to the identified environment ; and the processing the call using the automated call processing system comprises interpreting speech data received from the user during the call us ing the automatic speech recognition system . The method can, for example , improve the functionality of the automatic speech recognition system in various different call conditions .

[0007] In another implementation form of the first aspect , the configuring the automatic speech recognition system according to the identified environment comprises at least one of : configuring the automatic speech recognition system according to an amount of noise identified in the environment ; and/or configuring a filtering of the automatic speech recognition system according to the identified environment . The method can, for example , improve the functionality of the automatic speech recognition system in varying call noise conditions .

[0008] In another implementation form of the first aspect , the configuring the automatic speech recognition system according to the identified environment comprises selecting an automatic speech recognition instance used by the automatic speech recognition system according to the identified environment . The method can, for example , efficiently choose an appropriate automatic speech recognition instance for the call .

[0009] In another implementation form of the first aspect , processing the call using the automated call processing system comprises at least one of : configuring whether the automated call processing system repeats information during the call according to the identified environment ; configuring whether the automated call processing system sends information to the user using at least one communication channel other than the call according to the identified environment ; configuring whether the automated call processing system provides a call-back option to the user according to the identified environment ; configuring whether the automated call processing system forwards the call to a human according to the identified environment ; adj usting dialogue provided by the automated call processing system during the call according to the identified environment ; adj usting a priority of the call according to the identif ied environment ; and/or adj usting at least one property of a speaking voice provided by the automated call processing system during the call according to the identified environment . The method can, for example , efficiently configure at least some of the aforementioned properties of the automated call processing system .

[0010] In another implementation form of the first aspect , the at least one communication channel other than the call comprises at least one of : email , chat messaging, and/or text messaging . The method can, for example , efficiently send information to the user using alternative communication channels .

[001 1 ] In another implementation form of the first aspect , the adj usting the at least one property of the speaking voice provided by the automated call processing system during the call according to the identified environment comprises at least one of : selecting a recording or speech-synthesis voice according to the identified environment ; adj usting a speed of the speaking voice according to the identified environment ; and/or adj usting a volume of the speaking voice according to the identified environment . The method can, for example , improve the functionality of the automatic speech recognition system in varying call conditions .

[001 2] In another implementation form of the first aspect , the method further comprises transmitting information about the identified environment to another system . The method can, for example , allow other systems to utilise the information about the identified environment .

[001 3] In another implementation form of the first aspect , the identifying the environment of the user during the call using acoustic scene classification comprises identifying the environment of the user during the call using acoustic scene classification a plurality of times .

[0014] In another implementation form of the first aspect , the selecting the automatic speech recognition instance used by the automatic speech recognition system according to the identified environment comprises selecting an automatic speech recognition instance used by the automatic speech recognition system according to the identified environment a plurality of times .

[001 5] According to a second aspect , a computing device compri ses at least one processor and at least one memory including computer program code , the at least one memory and the computer program code being configured to , with the at least one proces sor, cause the computing device to perform the method according to the first aspect . [0016] According to a third aspect , a computer program product comprises program code configured to perform the method according to the first aspect when the computer program product is executed on a computer .

[001 7] Many of the attendant features wil l be more readily appreciated as they become better understood by reference to the following detailed description considered in connection with the accompanying drawings .

DESCRIPTION OF THE DRAWINGS

[0018] In the following, example embodiments are described in more detail with reference to the attached figures and drawings , in which :

[0019] Fig . 1 illustrates a flow chart representation of a method according to an embodiment ;

[0020] Fig . 2 illustrates a signalling diagram according to an embodiment ;

[0021 ] Fig . 3 illustrates a schematic representation of system modules according to an embodiment ;

[0022] Fig . 4 illustrates a schematic representation of modules of an automatic speech recognition system according to an embodiment ;

[0023] Fig . 5 illustrates a schematic representation of acoustic scene classification according to an embodiment ; and

[0024] Fig . 6 illustrates a schematic representation of an automatic speech recognition system according to an embodiment ; [0025] Fig . 7 illustrates a schematic representation of a computing device according to an embodiment .

[0026] In the following, like reference numerals are used to des ignate li ke parts in the accompanying drawings .

DETAILED DESCRIPTION

[0027] In the following description, reference is made to the accompanying drawings , which form part of the disclosure , and in which are shown, by way of illustration, specific aspects in which the present disclosure may be placed . It is understood that other aspects may be utilised, and structural or logical changes may be made without departing from the scope of the present disclosure . The following detailed description, therefore , is not to be taken in a limiting sense , as the scope of the present disclosure is defined be the appended claims .

[0028] For instance , it is understood that a disclosure in connection with a described method may also hold true for a corresponding device or system configured to perform the method and vice versa . For example , if a specific method step is described, a corresponding device may include a unit to perform the described method step, even if such unit is not explicitly described or il lustrated in the f igures . On the other hand, for example , if a specific apparatus is described based on functional units , a corresponding method may include a step performing the described functionality, even if such step is not explicitly described or illustrated in the figures . Further, it is understood that the features of the various example aspects described herein may be combined with each other, unless specifically noted otherwise .

[0029] Fig . 1 illustrates a flow chart representation of a method according to an embodiment .

[0030] According to an embodiment , a computer-implemented method 100 for automated call processing comprises receiving 101 a call from a user .

[0031 ] The call may also be referred to as a phone call , a voice call , or similar .

[0032] The method 100 may further comprise identifying

102 an environment of the user during the call using acoustic scene classification (ASC) .

[0033] The environment may also be referred to as an ASC class , a context , an acoustic scene , a location, or similar .

[0034] The method 100 may further comprise configuring

103 at least one property of an automated call processing system according to the identified environment . [0035] The at least one property of an automated call processing system may comprise , for example , one or more of the properties disclosed herein . The at least one property may also be referred to as at least one configuration or similar .

[0036] The automated call processing system may also be referred to as a call proces sing system, an automatic call processing system, a voicebot , a voicebot system, a call processing device , or similar .

[0037] The method 100 may further comprise processing 104 the call at least partially using the automated cal l processing system .

[0038] The automated call processing system may comprise , for example , a script according to which the system processes the call . The script may comprise , for example , questions that the automated call processing system as ks the user and how the system should respond to different answers provided by the user . The automated call processing system can comprise definition for various environments according to which the behaviour of the automated call processing system can change . For example , the script can comprise definitions of the form " if environment is X then perform action Y" .

[0039] There are various situations where a user calling a service provider using an automated call process ing system may be in a less than ideal environment or situation to interact with the automated call processing system . For example , the user may be in a noisy environment and may have difficulties hearing clearly, the background noise may negatively impact the speech recognition capabilities of the automated call processing system, the user may be on the move , the user may be driving, and/or the user has may not have access to a computer .

[0040] Also , the circumstances and the situation of the user cal ling the service provider may vary and the automated call processing system may not function in an optimal way in all cases . For example , one user may call and want to reserve train tickets well in advance while at home , whereas another user may want to reserve tickets while already rushing to the train station .

[0041 ] A service provider may correspond to , for example , an entity utilising the method . For example , the service provider may be a company and the user may be a customer of that company .

[0042] According to an embodiment , the automated call processing system comprises an automatic speech recognition (ASR) system and the configuring the at least one property of the automated call processing system comprises configuring the automatic speech recognition system according to the identified environment , and the processing the call using the automated call process ing system comprises interpreting speech data received from the user during the call using the automatic speech recognition system .

[0043] The automated call processing system may comprise the ASR system and/or the ASC system or the ASR system and/or the ASC system can be separate systems from the automated call processing system . The ASC system may also be referred to as an ASC, an ASC module , an ASC function, or similar . The ASR system may also be referred to as an ASR, an ASR module , an ASR function, or similar . [0044] I f the call is forwarded to a human, the human can, for example , as ses s the situation and process the call at least partially manually .

[0045] According to an embodiment , the configuring the automatic speech recognition system according to the identified environment comprises at least one of : configuring the automatic speech recognition system according to an amount of noise identi fied in the environment and/or configuring a filtering of the automatic speech recognition system according to the identified environment .

[0046] The automated call processing system may also apply filtering to the audio signal of the call or inform the ASR to apply fi ltering based on the identified environment .

[0047] According to an embodiment , the configuring the automatic speech recognition system according to the identified environment comprises selecting an automatic speech recognition instance used by the automatic speech recognition system according to the identified environment .

[0048] An ASR instance may refer to a specific way of configuring the ASR system . For example , each ASR instance may be optimi zed for specific audio characteristics , such as noi se level . The ASR system may comprise a plurality of ASR instances according to which the ASR system can be configured . An ASR instance may also be referred to as an ASR configuration, an ASR, or similar . [0049] The identified environment can impact the selection of the ASR instance that the automated call processing system utili zes . For example , the automated cal l processing system may select an ASR instance that has been optimi zed for noisy speech when the identified environment comprises , for example , a street , a public place , the outdoors , or similar .

[0050] According to an embodiment , the identifying the environment of the user during the call us ing acoustic scene classification comprises identifying the environment of the user during the cal l using acoustic scene classification a plurality of times .

[0051 ] For example , acoustic scene classification may be performed for each time period in a plurality of time periods .

[0052] According to an embodiment , the selecting the automatic speech recognition instance used by the automatic speech recognition system according to the identified environment comprises selecting an automatic speech recognition instance used by the automatic speech recognition system according to the identified environment a plurality of times .

[0053] For example , an automatic speech recognition instance used by the automatic speech recognition system may be selected for each time period in a plural ity of time periods .

[0054] For example , during the speech recognition process , the ASC may perform scene classification multiple times or substantially continuously so that if the acoustic scene changes during the call due to , for example , the cal ler moving from outdoors to indoors , the ASR may be changed accordingly .

[0055] According to an embodiment , the configuring the automatic speech recognition system according to the identified environment comprises selecting an ASR used by the automatic speech recognition system from a plurality of available ASRs according to the identified environment . The selected ASR can comprise , for example , an optimal ASR for the identified environment .

[0056] In some embodiments , the ASC may be applied to the call or other voice samples that needs to be transcribed .

[0057] According to an embodiment , processing the call using the automated call processing system comprises configuring whether the automated call processing system repeats information during the cal l according to the identified environment .

[0058] The automated call processing system may repeat information and/or ask for confirmation if , for example , the identified environment comprises a noisy environment where the user may have dif ficulties to hear the voice provided by the automated call proces sing system . In silent environments , such repetition or requesting confirmation may not be needed and may be omitted in order to make the dialogue between the user and the automated call processing system more fluent .

[0059] Additionally or alternatively, the processing the call using the automated call processing system may comprise configuring whether the automated call processing system sends information to the user using at least one communication channel other than the call according to the identified environment .

[0060] Additionally or alternatively, the processing the call using the automated call processing system may comprise configuring whether the automated call processing system provides a call-back option to the user according to the identified environment .

[0061 ] Additionally or alternatively, the processing the call using the automated call processing system may comprise configuring whether the automated call processing system forwards the call to a human according to the identified environment .

[0062] Additionally or alternatively, the processing the call using the automated call processing system may comprise adj usting dialogue provided by the automated call processing system during the call according to the identified environment .

[0063] Additionally or alternatively, the processing the call using the automated call processing system may comprise adj usting a priority of the call according to the identified environment .

[0064] The automated call processing system may adj ust the priority of the user ' s issue . For example , a user making a train ticket reservation in a train station may be prioriti zed, since the issue is probably urgent . [0065] Additionally or alternatively, the processing the call using the automated call processing system comprises adj usting at least one property of a speaking voice provided by the automated call processing system during the call according to the identified environment . [0066] According to an embodiment , the at least one communication channel other than the call comprises at least one of : email , chat messaging, and/or text messaging .

[0067] For example , the automated call processing system may send information via emai l , chat messaging, and/or text messaging to a user on the move .

[0068] According to an embodiment , the adj usting the at least one property of the speaking voice provided by the automated call processing system during the call according to the identified environment comprises selecting a recording or speech-synthesis voice according to the identified environment .

[0069] The automated call processing system may adj ust the speaking voice provided to the user during the call according to the identified environment . The adj usting may comprise , for example , selecting recordings and/or speech-synthesis voice so that it is , for example , clearer or slower for environments with background noise .

[0070] Additionally or alternatively, the adj usting the at least one property of the speaking voice provided by the automated call processing system during the call according to the identified environment may comprise adj usting a speed of the speaking voice according to the identified environment .

[0071 ] Additionally or alternatively, the adj usting the at least one property of the speaking voice provided by the automated call processing system during the call according to the identified environment may comprise adj usting a volume of the speaking voice according to the identified environment .

[0072] The volume of speech provided by the automated call processing system, such as announcements , prompts , etc . , can be adj usted according to the background noise volume in the call .

[0073] At least some embodiments disclosed herein can improve user experience with the automated call processing system .

[0074] At least some embodiments disclosed herein can improve efficiency of the automated call processing system . Thus , more call may be handled automatically by the automated call processing system .

[0075] At least some embodiments disclosed herein can improve ASR accuracy .

[0076] The various processing operations disclosed herein may be performed in various different orders . For example , the embodiment of Fig . 1 only illustrates an exemplary order of operations . Furthermore , at least some operations may be performed at least partially in parallel .

[0077] Fig . 2 illustrates a signalling diagram according to an embodiment . [0078] A user 201 can call 205 the automated call processing system 202 .

[0079] The automated call processing system 202 can forward 206 the audio of the call to an ASC system 203 for analysis .

[0080] The ASC system 203 can return information about the identified environment 207 to the automated call processing system 202 .

[0081 ] The automated call processing system 202 can respond 208 to the user 201 adj usting its response based on the identified environment .

[0082] The user 201 can respond 209 to the automated call processing system 202 .

[0083] The automated call processing system 202 can select the ASR 204 to be used based on identified environment and forward 210 audio of the call to the selected ASR 204 .

[0084] The ASR 204 can transcribe the speech of the call and return the transcript 211 to the automated cal l processing system 202 . The transcript 211 may comprise text data corresponding to the audio of the call .

[0085] The automated call processing system 202 can use the transcript 211 to respond 212 to the user 201 and may adj ust the response based on identified environment .

[0086] Operations 209 - 212 may be repeated as needed during the call . In some situations , operations 206 and 207 may also be repeated, for example , periodically in order to detect if the environment changes . [0087] Fig . 3 illustrates a schematic representation of system modules according to an embodiment .

[0088] The call 205 may be provided via, for example , a telephony network 301 , such as a public switched telephone network ( PSTN) or a mobile telephone network, voice over I P (VoI P) network, or similar .

[0089] A script 302 can define what the automated call processing system 202 should do and how to behave during the call . The script 302 can also comprise information about how to deal with various identi fied environments . [0090] The automated call processing system 202 can be responsible for the overall control and orchestration of the dialogue with the user 201 . The automated cal l processing system 202 can perform the defined script 302 .

[0091 ] The ASR system 204 can perform the actual speech-to-text conversion . There can be multiple ASR instances that the automated call processing system 202 may utili ze .

[0092] The ASC system 203 can receive audio from the user 201 via the automated call processing system 202 and return the identified environment . Additionally, the ASC system 203 can return a proposal for the best ASR to be used . The system 202 can select the ASR to be used based on input from the ASC 203 .

[0093] According to an embodiment , the method further comprises transmitting information about the identified environment to another system . [0094] Customer Relationship Management (CRM) and Enterprise Resource Planning (ERP) 303 are examples of other systems that may receive information about the identified environment . The automated call processing system 202 may send information about the identified environment to an external system such as a CRM or an ERP 303 .

[0095] Fig . 4 illustrates a schematic representation of modules of an automatic speech recognition system according to an embodiment .

[0096] ASR systems can utili ze principles from several different fields , such as signal processing, artificial intelligence , and linguistics , in order to automatically convert an audio signal comprising speech into text that corresponds to the content of the speech in the system' s input signal . An embodiment of an ASR system is il lus trated in Fig . 4 .

[0097] An ASR system can perform feature extraction

401 on speech data 410 . The extracted features can be provided to an acoustic model 402 . The acoustic model

402 can comprise a statistical model that identifies sound units from an input speech signal 410 after relevant features have been extracted from it .

[0098] A decoder 405 can deduce the text based on information from various components , such as the acoustic model 402 , a language model 403 , and a lexicon 404 . The language model 403 can comprise a statistical model that scores how likely words are to occur with each other in a given language . The lexicon 404 can comprise a pronunciation dictionary that indicates how words are constructed from sound units .

[0099] The acoustic model 402 may be produced using audio-text pairs where the text part corresponds to the speech in the audio signal . The language model 403 can be produced using textual resources for the target language , like English, while the lexicon 404 can be created with the help of linguists .

[0100] The embodiment of Fig . 4 is only an example of an ASR system . Alternatively, the ASR system can be implemented in various alternative ways .

[0101 ] Fig . 5 illustrates a schematic representation of acoustic scene classification according to an embodiment .

[0102] ASC can classify audio samples to categories corresponding to the environment in which the audio sample was recorded . For instance , an audio sample could be recorded in an office , in an airport , or in a factory . One of the obj ectives of ASC can be to provide contextawareness to automated audio systems .

[0103] The input of an ASC pipeline is an audio signal 501 . Before performing the ASC operation per se , a feature extraction step 502 can be applied to the audio signal 501 . This step can trans form the audio signal 501 into a format that contains relevant information for the actual classification operation and could involve , for example , signal processing algorithms such as the Fast Fourier Transform ( FFT ) algorithm . [0104] Once the feature extraction step 502 has been performed, the extracted features are used to perform the actual ASC operation 503 . The ASC module can be a statistical model that assigns the most likely category (environment /scene) based on the input features . In the embodiment of Fig . 5 , the selected acoustic scene 504 is "Office" . The statistical module is typically produced using a training step by using audio-context pairs that can be collected, for instance , via human annotators .

[0105] In some embodiments , the method 100 may be implemented using various modules /components such as those disclosed herein . Alternatively, the method 100 may also be implemented using various other systems .

[0106] Fig . 6 illustrates a schematic representation of an automatic speech recognition system according to an embodiment .

[0107] The embodiment of Fig . 6 illustrates a schematic representation of a so-called end-to-end (E2E ) ASR architecture . An ASR architecture can comprise an end- to-end model 610 which may comprise a neural network architecture i . e . a trainable model that can be trained to output text 611 based on a speech input 410 using audio-transcript pairs . Thus , for these types of architectures , the selection of an acoustic model may not be relevant as they do not comprise a standalone acoustic model contrary to , for example , an architecture illustrated in the embodiment of Fig . 4 . Instead, the roles of the different traditional components can be learned by the single neural network architecture. Thus, the training procedure of such systems can be simplified.

[0108] In any embodiment disclosed herein, an ASR, an ASR system, and/or an ASR instance can comprise, for example, end-to-end ASR architecture, an ASR architecture similar to that disclosed in the embodiment of Fig. 4, or any other type of ASR architecture.

[0109] Fig. 7 illustrates a schematic representation of a computing device according to an embodiment.

[0110] According to an embodiment, a computing device 600 comprises at least one processor 601 and at least one memory 602 including computer program code, the at least one memory 602 and the computer program code configured to, with the at least one processor 601, cause the computing device to perform the method 100.

[0111] The computing device 600 may comprise at least one processor 601. The at least one processor 601 may comprise, for example, one or more of various processing devices, such as a co-processor, a microprocessor, a digital signal processor (DSP) , a processing circuitry with or without an accompanying DSP, or various other processing devices including integrated circuits such as, for example, an application specific integrated circuit (ASIC) , a field programmable gate array (FPGA) , a microprocessor unit (MCU) , a hardware accelerator, a special-purpose computer chip, or the like.

[0112] The computing device 600 may further comprise a memory 602. The memory 602 may be configured to store, for example, computer programs and the like. The memory 602 may comprise one or more volatile memory devices, one or more non-volatile memory devices, and/or a combination of one or more volatile memory devices and nonvolatile memory devices. For example, the memory 602 may be embodied as magnetic storage devices (such as hard disk drives, magnetic tapes, etc.) , optical magnetic storage devices, and semiconductor memories (such as mask ROM, PROM (programmable ROM) , EPROM (erasable PROM) , flash ROM, RAM (random access memory) , etc.) .

[0113] The computing device 600 may further comprise other components not illustrated in the embodiment of Fig. 7. The computing device 600 may comprise, for example, an input/output bus for connecting the computing device 600 to other devices. Further, a user may control the computing device 600 via the input/output bus.

[0114] When the computing device 600 is configured to implement some functionality, some component and/or components of the computing device 600, such as the at least one processor 601 and/or the memory 602, may be configured to implement this functionality. Furthermore, when the at least one processor 601 is configured to implement some functionality, this functionality may be implemented using program code comprised, for example, in the memory.

[0115] The computing device 600 may be implemented at least partially using, for example, a computer, some other computing device, or similar. [01 16] The method 100 and/or the computing device 600 may be utili sed in, for example , in a so-called voice- bot . A voicebot may be configured to obtain information from users by, for example, phone and convert the voice information into text information using ASR . The method 100 may be used to improve functionality of the ASR . The voicebot may further be configured to further process , such as classify, the text information . The voicebot can, for example , as k questions about , for example , basic information from a customer in a customer service situation over the phone , obtain the answers using ASR and the method 100 , and save the information in a system . Thus , the customer service situation can be made more efficient and user experience can be improved .

[01 1 7] Any range or device value given herein may be extended or altered without losing the effect sought . Also any embodiment may be combined with another embodiment unless explicitly disallowed .

[01 18] Although the subj ect matter has been described in language specific to structural features and/or acts , it is to be understood that the subj ect matter defined in the appended claims is not necessarily limited to the specific features or acts described above . Rather, the specific features and acts described above are disclosed as examples of implementing the claims and other equivalent features and acts are intended to be within the scope of the claims .

[01 19] It will be understood that the benefits and advantages described above may relate to one embodiment or may relate to several embodiments . The embodiments are not limited to those that solve any or all of the stated problems or those that have any or all of the stated benefits and advantages . It wil l further be understood that reference to ' an ' item may refer to one or more of those items .

[01 20] The steps of the methods described herein may be carried out in any suitable order, or simultaneously where appropriate . Additionally, individual blocks may be deleted from any of the methods without departing from the spirit and scope of the subj ect matter described herein . Aspects of any of the embodiments described above may be combined with aspects of any of the other embodiments described to form further embodiments without losing the effect sought .

[01 21 ] The term ' comprising ' is used herein to mean including the method, blocks or elements identified, but that such blocks or elements do not comprise an exclusive list and a method or apparatus may contain additional blocks or elements .

[01 22] It will be understood that the above description is given by way of example only and that various modif ications may be made by those s kil led in the art . The above specification, examples and data provide a complete description of the structure and use of exemplary embodiments . Although various embodiments have been described above with a certain degree of particularity, or with reference to one or more individual embodiments , those skilled in the art could make numerous alterations to the disclosed embodiments without departing from the spirit or scope of this specification .