CN112235468A - Audio processing method and system for voice customer service evaluation - Google Patents
Audio processing method and system for voice customer service evaluation Download PDFInfo
- Publication number
- CN112235468A CN112235468A CN202011106805.6A CN202011106805A CN112235468A CN 112235468 A CN112235468 A CN 112235468A CN 202011106805 A CN202011106805 A CN 202011106805A CN 112235468 A CN112235468 A CN 112235468A
- Authority
- CN
- China
- Prior art keywords
- customer service
- audio
- candidate
- content
- complaint
- 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.)
- Withdrawn
Links
- 238000011156 evaluation Methods 0.000 title claims abstract description 31
- 238000003672 processing method Methods 0.000 title claims abstract description 26
- 238000000034 method Methods 0.000 claims abstract description 19
- 238000012216 screening Methods 0.000 claims abstract description 5
- 238000000605 extraction Methods 0.000 claims description 29
- 238000012545 processing Methods 0.000 claims description 11
- 238000001514 detection method Methods 0.000 claims description 10
- 230000001755 vocal effect Effects 0.000 claims description 7
- 238000004458 analytical method Methods 0.000 claims description 4
- 238000013441 quality evaluation Methods 0.000 abstract description 7
- 239000000284 extract Substances 0.000 description 9
- 230000014509 gene expression Effects 0.000 description 8
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000008451 emotion Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M3/00—Automatic or semi-automatic exchanges
- H04M3/22—Arrangements for supervision, monitoring or testing
- H04M3/2227—Quality of service monitoring
-
- 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
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
- G10L25/63—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination for estimating an emotional state
-
- 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
- H04M3/50—Centralised arrangements for answering calls; Centralised arrangements for recording messages for absent or busy subscribers ; Centralised arrangements for recording messages
- H04M3/51—Centralised call answering arrangements requiring operator intervention, e.g. call or contact centers for telemarketing
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Signal Processing (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- Computational Linguistics (AREA)
- Physics & Mathematics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- Hospice & Palliative Care (AREA)
- Child & Adolescent Psychology (AREA)
- Quality & Reliability (AREA)
- Business, Economics & Management (AREA)
- Marketing (AREA)
- Telephonic Communication Services (AREA)
Abstract
The invention relates to an audio processing method and system for voice customer service evaluation, wherein the method comprises the steps of obtaining the customer service audio of each customer service staff according to the ID of the customer service staff, and screening to obtain candidate customer service audio according to the time length of the customer service audio; judging whether the candidate customer service audio comprises complaint contents or not by adopting a preset complaint content judgment condition according to the voice text, the sound frequency and the sound volume of the candidate customer service audio; and judging whether the candidate customer service audio comprises the suggestive content or not by adopting a preset suggestive content judgment condition according to the voice text, the sound frequency and the volume of the candidate customer service audio, and storing the corresponding relation between the candidate customer service audio and the ID of the customer service personnel in a complaining database or a suggestive database when the complaining content or the suggestive content exists. By adopting the method and the device, the customer service audio of the customer service personnel can be automatically processed, and the working quality evaluation efficiency of the voice customer service personnel is improved.
Description
Technical Field
The invention relates to the technical field of internet, in particular to an audio processing method and system for voice customer service evaluation.
Background
With the rapid development of society, the development of service industry is also changing day by day. In order to provide better service for clients, more and more industries have some customer service staff responsible for answering incoming calls of clients and solving corresponding problems for the clients, for example, the customer service staff in the property industry can handle matters such as house maintenance, business consultation and the like, the customer service staff in a travel platform can handle matters such as travel booking, travel recommendation and hotel booking, and the customer service staff in the insurance industry can handle matters such as insurance application consultation and refund saving. Therefore, the main task of the customer service staff is to explain different problems to different customers, and usually, the customer only calls the customer service when a problem occurs.
The quality of service of the customer service personnel directly affects the look and feel of the customer on the brand or enterprise. Therefore, it is very important to evaluate the quality of the service of the customer service staff. The existing customer service quality evaluation is generally carried out through manual subjective judgment, so that the evaluation is not accurate enough, the efficiency is low, and time and labor are wasted.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides an audio processing method for voice customer service evaluation, which can automatically process the customer service audio of a customer service worker and improve the working quality evaluation efficiency of the voice customer service worker.
In order to achieve the above object, the present invention has the following configurations:
the invention provides an audio processing method for voice customer service evaluation, which comprises the following steps:
for each customer service person, acquiring the customer service audio of the customer service person according to the ID of the customer service person;
selecting audio time length larger than a preset time length threshold value from the customer service audio to obtain candidate customer service audio;
extracting voice from the candidate customer service audio, and performing text recognition on the voice to obtain a voice text of the candidate customer service audio;
detecting the sound frequency and the sound volume in the candidate customer service audio to obtain the sound frequency and the sound volume of the candidate customer service audio;
judging whether the candidate customer service audio comprises complaint content or not by adopting a preset complaint content judgment condition according to the voice text, the sound frequency and the sound volume of the candidate customer service audio, and if yes, storing the corresponding relation between the complaint content and the ID of the customer service staff in a complaint database;
and judging whether the candidate customer service audio comprises the showing content or not by adopting a preset showing content judgment condition according to the voice text, the sound frequency and the volume of the candidate customer service audio, and if so, storing the corresponding relation between the showing content and the ID of the customer service personnel in a showing database.
Optionally, the determining, by using a preset complaint content determining condition, whether the candidate customer service audio includes a complaint content includes the following steps:
judging whether the candidate customer service audio comprises a preset complaint keyword or not;
if yes, judging whether the sound frequency of the part, corresponding to the complaint keyword, of the candidate customer service audio is larger than a first frequency threshold and the sound volume is larger than a first sound volume threshold, if yes, the candidate customer service audio comprises complaint contents, and if not, the candidate customer service audio does not comprise the complaint contents;
if not, then the candidate customer service audio does not include complaint content.
Optionally, the determining whether the candidate customer service audio includes a preset complaint keyword includes the following steps:
carrying out voiceprint detection on the candidate customer service audio, and distinguishing a customer service audio part and a client audio part;
and judging whether the client audio part in the candidate client service audio comprises a preset complaint keyword or not.
Optionally, the determining whether the candidate customer service audio includes the suggestive content by using a preset suggestive content determining condition includes the following steps:
judging whether the candidate customer service audio comprises a preset raise keyword or not;
if yes, judging whether the sound frequency of the part, corresponding to the showing keyword, in the candidate customer service audio is larger than a second frequency threshold, and the sound volume is larger than a second sound volume threshold, if yes, the candidate customer service audio comprises showing content, and if not, the candidate customer service audio does not comprise the showing content;
and if not, the candidate customer service audio does not comprise the showing content.
Optionally, the determining whether the candidate customer service audio includes a preset raise keyword includes the following steps:
carrying out voiceprint detection on the candidate customer service audio, and distinguishing a customer service audio part and a client audio part;
and judging whether the client audio part in the candidate customer service audio comprises a preset raised keyword or not.
Optionally, after obtaining the speech text of the candidate customer service audio, the method further includes the following steps:
judging whether the number of the customer service audios is larger than a preset audio number threshold value or not;
if yes, detecting a classification keyword in the voice text of the candidate customer service audio, and classifying the candidate customer service audio;
selecting candidate customer service audios of a preset specified category from the candidate customer service audios, and taking the candidate customer service audios of other categories as the candidate customer service audios no longer;
after judging whether the candidate customer service audio comprises the complaint content, the method further comprises the following steps: and if the complaint content exists, storing the complaint content and the audio category corresponding to the complaint content in a complaint database.
Optionally, the method further includes, for each customer service, determining a specified category corresponding to the customer service by:
acquiring the complaint content in the preset time period before the customer service from the complaint database according to the ID of the customer service staff;
and counting the audio category with the most complaint times in the audio categories corresponding to the complaint content before the customer service as a preset specified category of the customer service.
Optionally, for each customer service, determining a specified category corresponding to the customer service further includes the following steps:
judging whether the occurrence frequency of the complaint content in a preset time period before the customer service is greater than a preset frequency threshold value or not according to the ID of the customer service staff;
if yes, obtaining the complaint content in the preset time period before the customer service from the complaint database according to the ID of the customer service staff, and counting the audio category with the most complaint times in the audio categories corresponding to the complaint content before the customer service as the preset specified category of the customer service;
if not, acquiring the group information of the customer service staff according to the ID of the customer service staff, inquiring the complaint content of other customer service staff in the same group of the customer service staff in a previous preset time period from the complaint database according to the group information, and counting the audio category with the largest number of complaints as a preset specified category of the customer service.
An audio processing system for voice customer service evaluation according to an embodiment of the present invention is applied to the audio processing method for voice customer service evaluation, and the system includes:
the audio acquisition module is used for acquiring the customer service audio of each customer service staff according to the ID of the customer service staff;
the audio screening module is used for selecting audio time length which is greater than a preset time length threshold value from the customer service audio to obtain candidate customer service audio;
the voice analysis module is used for extracting voice from the candidate customer service audio and performing text recognition on the voice to obtain a voice text of the candidate customer service audio; the voice frequency and the volume of the candidate customer service audio are detected to obtain the voice frequency and the volume of the candidate customer service audio;
the content extraction module is used for judging whether the candidate customer service audio comprises complaint content or not by adopting a preset complaint content judgment condition according to the voice text, the sound frequency and the volume of the candidate customer service audio, and storing the corresponding relation between the complaint content and the ID of the customer service staff in a complaint database if the candidate customer service audio comprises the complaint content; and judging whether the candidate customer service audio comprises the showing content or not by adopting a preset showing content judgment condition according to the voice text, the sound frequency and the volume of the candidate customer service audio, and if so, storing the corresponding relation between the showing content and the ID of the customer service personnel in a showing database.
Optionally, the content extraction module adopts a preset complaint content judgment condition to judge whether the candidate customer service audio includes a complaint content by the following steps:
the content extraction module judges whether the candidate customer service audio comprises a preset complaint keyword or not;
if yes, the content extraction module judges whether the sound frequency of the part, corresponding to the complaint keyword, of the candidate customer service audio is greater than a first frequency threshold and the sound volume is greater than a first sound volume threshold, if yes, the candidate customer service audio comprises the complaint content, and if not, the candidate customer service audio does not comprise the complaint content;
if not, then the candidate customer service audio does not include complaint content;
the content extraction module adopts the following steps to adopt a preset suggestive content judgment condition to judge whether the candidate customer service audio comprises suggestive content:
the content extraction module judges whether the candidate customer service audio comprises a preset raise keyword or not;
if yes, the content extraction module judges whether the sound frequency of the part, corresponding to the showing keyword, in the candidate customer service audio is larger than a second frequency threshold, and the volume is larger than a second volume threshold, if yes, the candidate customer service audio comprises the showing content, otherwise, the candidate customer service audio does not comprise the showing content;
and if not, the candidate customer service audio does not comprise the showing content.
In summary, compared with the prior art, the audio processing method and system for voice customer service evaluation provided by the invention can automatically process the customer service audio of the customer service staff, automatically extract the keywords included in the audio, judge whether the audio relates to the customer complaint or the customer raise, and store the corresponding relationship between the content of the complaint or the raise and the ID of the customer service staff in the complaint database or the raise database when the customer complaint or the customer raise occurs, so that the audio processing method and system can be conveniently and directly used when the customer service staff subsequently evaluates the work quality, thereby improving the work quality evaluation efficiency of the voice customer service staff, being beneficial to improving the service quality of the customer, and further improving the enterprise brand value.
Drawings
FIG. 1 is a schematic diagram of an audio processing method for voice customer service evaluation according to an embodiment of the present invention;
FIG. 2 is a flowchart of one embodiment of the present invention for determining whether the candidate customer service audio includes complaint content;
FIG. 3 is a flowchart of determining whether the candidate customer service audio includes a suggestive content, in accordance with an embodiment of the present invention;
fig. 4 is a schematic structural diagram of an audio processing system for voice customer service evaluation according to an embodiment of the present invention.
Detailed Description
Example embodiments will now be described more fully with reference to the accompanying drawings. Example embodiments may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of example embodiments to those skilled in the art. The same reference numerals in the drawings denote the same or similar structures, and thus their repetitive description will be omitted.
The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to provide a thorough understanding of embodiments of the invention. One skilled in the relevant art will recognize, however, that the invention may be practiced without one or more of the specific details, or with other methods, components, materials, and so forth. In some instances, well-known structures, materials, or operations are not shown or described in detail to avoid obscuring the invention.
As shown in fig. 1, to solve the technical problems in the prior art, the present invention provides an audio processing method for voice customer service evaluation, comprising the following steps:
for each customer service person, acquiring the customer service audio of the customer service person according to the ID of the customer service person, wherein the voice call between the customer service person and a customer is recorded completely, so that the recorded customer service audio of the customer service person can be directly acquired, wherein the customer audio is the audio of a complete voice call service process, and comprises the voice of the customer and the voice of the customer service;
considering that the customer service audio with a very short duration has a very small reference value, the customer service audio is further simply screened according to the duration, the subsequent audio processing amount is reduced, and the audio processing efficiency is improved, specifically, the customer service audio with the audio duration larger than a preset duration threshold is selected from the customer service audio to obtain candidate customer service audio, and the numerical value of the preset duration threshold can be selectively set according to needs, such as 1 minute, 2 minutes and the like;
extracting voice from the candidate customer service audio, and performing text recognition on the voice to obtain a voice text of the candidate customer service audio, wherein the conversion from the voice to the text can be realized by adopting a voice recognition technology in the prior art;
detecting the sound frequency and the sound volume in the candidate customer service audio to obtain the sound frequency and the sound volume of the candidate customer service audio;
judging whether the candidate customer service audio comprises complaint content or not by adopting a preset complaint content judgment condition according to the voice text, the sound frequency and the sound volume of the candidate customer service audio, and if yes, storing the corresponding relation between the complaint content and the ID of the customer service staff in a complaint database, wherein the complaint content refers to the voice text corresponding to the complaint content or a keyword of the voice text;
and judging whether the candidate customer service audio comprises the suggestive content or not by adopting a preset suggestive content judgment condition according to the voice text, the voice frequency and the volume of the candidate customer service audio, and if so, storing the corresponding relation between the suggestive content and the ID of the customer service personnel in a suggestive database, wherein the suggestive content refers to the voice text corresponding to the suggestive content or the keyword of the voice text.
Therefore, the invention provides an audio processing method for voice customer service evaluation, which can automatically process the customer service audio of the customer service staff, automatically extract keywords included in the audio, judge whether the audio relates to customer complaints or customer expressions, and store the corresponding relation between the content of the complaints or expressions and the ID of the customer service staff in a complaint database or an expression database when the customer complaints or the customer expressions occur, so that the audio processing method is convenient to be directly used when the work quality of the customer service staff is subsequently evaluated, and the work quality evaluation efficiency of the voice customer service staff is improved.
As shown in fig. 2, in this embodiment, the determining whether the candidate customer service audio includes the complaint content by using the preset complaint content determining condition includes the following steps:
judging whether the candidate customer service audio comprises preset complaint keywords, wherein the complaint keywords can be preset to comprise complaint keywords, errors keywords, unresolved keywords and the like;
if yes, judging whether the sound frequency of the part of the candidate customer service audio corresponding to the complaint keyword is larger than a first frequency threshold value and the sound volume is larger than a first sound volume threshold value, if yes, indicating that the emotion of the user is more excited when speaking the part related to the complaint keyword, wherein the candidate customer service audio comprises complaint content, otherwise, the candidate customer service audio does not comprise the complaint content;
if not, then the candidate customer service audio does not include complaint content.
Further, the complaint keyword may include a plurality of combinations, for example, "multiple times", "unresolved", "again", and the like as a complaint keyword combination, and when it is determined whether the candidate customer service audio includes a preset complaint keyword, if the candidate customer service audio includes only one word of the complaint keyword combination, for example, "multiple times", it may not be possible to indicate that the customer is a complaint, and it is required that the audio of the customer includes more than two or even all of the complaint keywords in the complaint keyword combination to determine that the candidate customer service audio includes the preset complaint keyword, so as to improve the accuracy of complaint identification.
In this embodiment, because the speech part of the customer is automatically analyzed, the determination and extraction of the complaint content only need to be performed on the audio part of the customer, specifically, the determination of whether the candidate customer service audio includes the preset complaint keyword includes the following steps:
performing voiceprint detection on the candidate customer service audio to distinguish a customer service audio part from a client audio part, wherein the voiceprint detection only needs to identify the voiceprint of the customer service, and the part which does not conform to the voiceprint of the customer service is used as the client audio part;
and judging whether the client audio part in the candidate client service audio comprises a preset complaint keyword or not.
As shown in fig. 3, in this embodiment, the determining whether the candidate customer service audio includes the suggestive content by using a preset suggestive content determining condition includes the following steps:
judging whether the candidate customer service audio comprises preset raise keywords, wherein the preset raise keywords can comprise 'good', 'very professional', 'thank you' and the like;
if yes, judging whether the sound frequency of the part, corresponding to the showing keyword, in the candidate customer service audio is larger than a second frequency threshold, and the sound volume is larger than a second sound volume threshold, if yes, the candidate customer service audio comprises showing content, and if not, the candidate customer service audio does not comprise the showing content;
and if not, the candidate customer service audio does not comprise the showing content.
Further, the raise-word keyword may include a plurality of combinations, for example, "once", "fast", "thank you" and the like are used as one raise-word keyword combination, and when it is determined whether the candidate customer service audio includes the preset raise-word keyword, if the candidate customer service audio includes only one word of the raise-word keyword combination, for example, "once", it may not be possible to indicate that the customer is in raise, and it is required that the audio of the customer includes more than two, even all, of the raise-word keywords in the raise-word keyword combination, and it is determined that the candidate customer service audio includes the preset raise-word keyword, so as to improve accuracy of raise recognition.
In this embodiment, because the speech part of the client is automatically analyzed, the method only needs to determine and extract the vocational content of the audio part of the client, and specifically determines whether the candidate customer service audio includes a preset vocational keyword, which includes the following steps:
performing voiceprint detection on the candidate customer service audio to distinguish a customer service audio part from a client audio part, wherein the voiceprint detection only needs to identify the voiceprint of the customer service, and the part which does not conform to the voiceprint of the customer service is used as the client audio part;
and judging whether the client audio part in the candidate customer service audio comprises a preset raised keyword or not.
In this embodiment, after obtaining the speech text of the candidate customer service audio, the method further includes the following steps:
judging whether the number of the customer service audios is larger than a preset audio number threshold value or not;
if the number of the customer service audios is larger than a preset audio number threshold, the number of the customer service audios is large, in order to reduce audio processing capacity and ensure uniformity of evaluation bases among different customer services, certain screening needs to be further performed on the customer service audios, specifically, classification keywords in a voice text of the candidate customer service audios are detected, and the candidate customer service audios are classified; the numerical value of the preset audio quantity threshold value can be selected and set according to needs, and specifically can be selected according to the processing capacity of the system;
and selecting candidate customer service audios of preset appointed categories from the candidate customer service audios, and taking the candidate customer service audios of other categories as the candidate customer service audios.
After judging whether the candidate customer service audio comprises the complaint content, the method further comprises the following steps: and if the complaint content exists, storing the complaint content and the audio category corresponding to the complaint content in a complaint database for reference when a preset specified category is subsequently set.
In this embodiment, the audio processing method for voice customer service evaluation further includes, for each customer service, determining a specified category corresponding to the customer service by:
acquiring the complaint content in the preset time period before the customer service from the complaint database according to the ID of the customer service staff; the time length of the preset time period can be set according to needs, such as one month, one quarter, and the like;
since the audio category corresponding to each complaint content is stored in the complaint database, the audio category with the highest number of complaints in the audio categories corresponding to the complaint contents before the customer service can be counted as the preset specified category of the customer service. I.e. selecting the audio category that was weak before the customer service as the pre-set assigned category.
In this embodiment, for each customer service, determining the specified category corresponding to the customer service further includes the following steps:
judging whether the occurrence number of the complaint content in a preset time period before the customer service is greater than a preset number threshold value or not according to the ID of the customer service staff, namely if the occurrence number of the complaint content is few, the data of the customer service is not referential, and performing statistical analysis according to the complaint content of the same group of staff;
if yes, obtaining the complaint content in the preset time period before the customer service from the complaint database according to the ID of the customer service staff, and counting the audio category with the most complaint times in the audio categories corresponding to the complaint content before the customer service as the preset specified category of the customer service;
if not, acquiring the group information of the customer service staff according to the ID of the customer service staff, inquiring the complaint content of other customer service staff in the same group of the customer service staff in a previous preset time period from the complaint database according to the group information, and counting the audio category with the largest number of complaints as a preset specified category of the customer service.
As shown in fig. 4, an audio processing system for speech customer service evaluation according to an embodiment of the present invention is applied to the audio processing method for speech customer service evaluation, and the system includes:
the audio acquisition module 100 is configured to acquire, for each customer service person, a customer service audio of the customer service person according to an ID of the customer service person, where the customer service audio may be obtained by recording in real time during a call between the customer service and a client and is stored in an audio management database, the audio management database further stores an ID of the customer service person corresponding to each customer service audio, a recording time, a recording duration, and the like, and the audio acquisition module 100 may query a corresponding recording audio from the audio management database according to the ID of the customer service person;
the audio screening module 200 is configured to select an audio time length greater than a preset time length threshold from the customer service audio to obtain a candidate customer service audio, where a value of the preset time length threshold may be selectively set as required, for example, 1 minute, 2 minutes, and the like;
a voice analysis module 300, configured to extract features in the candidate customer service audio, specifically, the voice analysis module 300 extracts a voice from the candidate customer service audio, and performs text recognition on the voice to obtain a voice text of the candidate customer service audio; the voice frequency and the volume of the candidate customer service audio are detected to obtain the voice frequency and the volume of the candidate customer service audio;
the content extraction module 400 is configured to determine and extract a complaint content and a suggestive content, specifically, the content extraction module 400 determines whether the candidate customer service audio includes a complaint content according to a voice text, a sound frequency and a sound volume of the candidate customer service audio and by using a preset complaint content determination condition, where the complaint content refers to a voice text or a keyword of the voice text corresponding to the complaint content, and if the complaint content exists, the content extraction module 400 stores a correspondence between the complaint content and an ID of the customer service person in a complaint database; and the content extraction module 400 judges whether the candidate customer service audio includes the suggestive content or not by using a preset suggestive content judgment condition according to the voice text, the sound frequency and the volume of the candidate customer service audio, and if the candidate customer service audio includes the suggestive content, the content extraction module 400 stores the corresponding relationship between the suggestive content and the ID of the customer service person in a suggestive database, wherein the suggestive content refers to the voice text corresponding to the suggestive content or the keyword of the voice text.
Therefore, the invention provides an audio processing system for voice customer service evaluation, which can automatically process the customer service audio of a customer service staff, automatically extract keywords included in the audio, judge whether the audio relates to customer complaints or customer expressions, and store the corresponding relation between the content of the complaints or expressions and the ID of the customer service staff in a complaint database or an expression database when the customer complaints or the customer expressions occur, so that the audio processing system is convenient to be directly used when the work quality of the customer service staff is subsequently evaluated, and the work quality evaluation efficiency of the voice customer service staff is improved.
In this embodiment, the content extraction module 400 may determine whether complaint content is included by using the steps shown in fig. 2, and specifically, the content extraction module 400 determines whether the candidate customer service audio includes complaint content by using a preset complaint content determination condition by using the following steps:
the content extraction module 400 determines whether the candidate customer service audio includes a preset complaint keyword;
if yes, the content extraction module 400 determines whether the sound frequency of the part of the candidate customer service audio corresponding to the complaint keyword is greater than a first frequency threshold and the sound volume is greater than a first sound volume threshold, if yes, the candidate customer service audio includes the complaint content, otherwise, the candidate customer service audio does not include the complaint content;
if not, then the candidate customer service audio does not include complaint content;
the content extraction module 400 may determine whether the candidate customer service audio includes the suggestive content by adopting the steps shown in fig. 3, specifically, the content extraction module 400 determines whether the candidate customer service audio includes the suggestive content by adopting a preset suggestive content determination condition by adopting the following steps:
the content extraction module 400 determines whether the candidate customer service audio includes a preset raise keyword;
if yes, the content extraction module 400 determines whether the sound frequency of the portion of the candidate customer service audio corresponding to the vocational keyword is greater than a second frequency threshold and the volume is greater than a second volume threshold, if yes, the candidate customer service audio includes vocational content, otherwise, the candidate customer service audio does not include vocational content;
and if not, the candidate customer service audio does not comprise the showing content.
Further, in order to distinguish the client audio part and the client audio part in the client service audio well, the content extraction module 400 is further configured to perform voiceprint detection on the candidate client service audio, distinguish the client service audio part from the client audio part, only process the client audio part when the steps shown in fig. 2 are used to determine and extract the complaint content, only process the client audio part when the steps shown in fig. 3 are used to determine and extract the expressive content, the voiceprint detection only needs to identify the voiceprint of the client service, and the part that does not conform to the voiceprint of the client service is used as the client audio part.
In summary, compared with the prior art, the audio processing method and system for voice customer service evaluation provided by the invention can automatically process the customer service audio of the customer service staff, automatically extract the keywords included in the audio, judge whether the audio relates to the customer complaint or the customer raise, and store the corresponding relationship between the content of the complaint or the raise and the ID of the customer service staff in the complaint database or the raise database when the customer complaint or the customer raise occurs, so that the audio processing method and system can be conveniently and directly used when the customer service staff subsequently evaluates the work quality, thereby improving the work quality evaluation efficiency of the voice customer service staff, being beneficial to improving the service quality of the customer, and further improving the enterprise brand value.
In this specification, the invention has been described with reference to specific embodiments thereof. It will, however, be evident that various modifications and changes may be made thereto without departing from the broader spirit and scope of the invention. The specification and drawings are, accordingly, to be regarded in an illustrative rather than a restrictive sense.
Claims (10)
1. An audio processing method for voice customer service evaluation, the method comprising the steps of:
for each customer service person, acquiring the customer service audio of the customer service person according to the ID of the customer service person;
selecting audio time length larger than a preset time length threshold value from the customer service audio to obtain candidate customer service audio;
extracting voice from the candidate customer service audio, and performing text recognition on the voice to obtain a voice text of the candidate customer service audio;
detecting the sound frequency and the sound volume in the candidate customer service audio to obtain the sound frequency and the sound volume of the candidate customer service audio;
judging whether the candidate customer service audio comprises complaint content or not by adopting a preset complaint content judgment condition according to the voice text, the sound frequency and the sound volume of the candidate customer service audio, and if yes, storing the corresponding relation between the complaint content and the ID of the customer service staff in a complaint database;
and judging whether the candidate customer service audio comprises the showing content or not by adopting a preset showing content judgment condition according to the voice text, the sound frequency and the volume of the candidate customer service audio, and if so, storing the corresponding relation between the showing content and the ID of the customer service personnel in a showing database.
2. The audio processing method for voice customer service evaluation according to claim 1, wherein the step of judging whether the candidate customer service audio includes a content of complaint by using a preset condition for judging the content of complaint comprises the steps of:
judging whether the candidate customer service audio comprises a preset complaint keyword or not;
if yes, judging whether the sound frequency of the part, corresponding to the complaint keyword, of the candidate customer service audio is larger than a first frequency threshold and the sound volume is larger than a first sound volume threshold, if yes, the candidate customer service audio comprises complaint contents, and if not, the candidate customer service audio does not comprise the complaint contents;
if not, then the candidate customer service audio does not include complaint content.
3. The audio processing method for voice customer service evaluation according to claim 2, wherein the step of determining whether the candidate customer service audio includes a preset complaint keyword comprises the steps of:
carrying out voiceprint detection on the candidate customer service audio, and distinguishing a customer service audio part and a client audio part;
and judging whether the client audio part in the candidate client service audio comprises a preset complaint keyword or not.
4. The audio processing method for voice customer service evaluation according to claim 1, wherein the step of judging whether the candidate customer service audio includes a vocational content by using a preset vocational content judgment condition comprises the following steps:
judging whether the candidate customer service audio comprises a preset raise keyword or not;
if yes, judging whether the sound frequency of the part, corresponding to the showing keyword, in the candidate customer service audio is larger than a second frequency threshold, and the sound volume is larger than a second sound volume threshold, if yes, the candidate customer service audio comprises showing content, and if not, the candidate customer service audio does not comprise the showing content;
and if not, the candidate customer service audio does not comprise the showing content.
5. The audio processing method for voice customer service evaluation according to claim 4, wherein judging whether the candidate customer service audio includes a preset raise keyword comprises the following steps:
carrying out voiceprint detection on the candidate customer service audio, and distinguishing a customer service audio part and a client audio part;
and judging whether the client audio part in the candidate customer service audio comprises a preset raised keyword or not.
6. The audio processing method for speech customer service evaluation according to claim 1, further comprising the following steps after obtaining the speech text of the candidate customer service audio:
judging whether the number of the customer service audios is larger than a preset audio number threshold value or not;
if yes, detecting a classification keyword in the voice text of the candidate customer service audio, and classifying the candidate customer service audio;
selecting candidate customer service audios of a preset specified category from the candidate customer service audios, and taking the candidate customer service audios of other categories as the candidate customer service audios no longer;
after judging whether the candidate customer service audio comprises the complaint content, the method further comprises the following steps: and if the complaint content exists, storing the complaint content and the audio category corresponding to the complaint content in a complaint database.
7. The audio processing method for speech customer service evaluation according to claim 6, further comprising determining, for each customer service, a specified category to which the customer service corresponds using the steps of:
acquiring the complaint content in the preset time period before the customer service from the complaint database according to the ID of the customer service staff;
and counting the audio category with the most complaint times in the audio categories corresponding to the complaint content before the customer service as a preset specified category of the customer service.
8. The audio processing method for voice customer service evaluation according to claim 7, wherein said determining, for each customer service, the specified category to which the customer service corresponds further comprises the steps of:
judging whether the occurrence frequency of the complaint content in a preset time period before the customer service is greater than a preset frequency threshold value or not according to the ID of the customer service staff;
if yes, obtaining the complaint content in the preset time period before the customer service from the complaint database according to the ID of the customer service staff, and counting the audio category with the most complaint times in the audio categories corresponding to the complaint content before the customer service as the preset specified category of the customer service;
if not, acquiring the group information of the customer service staff according to the ID of the customer service staff, inquiring the complaint content of other customer service staff in the same group of the customer service staff in a previous preset time period from the complaint database according to the group information, and counting the audio category with the largest number of complaints as a preset specified category of the customer service.
9. An audio processing system for voice customer service evaluation, which is applied to the audio processing method for voice customer service evaluation according to any one of claims 1 to 8, the system comprising:
the audio acquisition module is used for acquiring the customer service audio of each customer service staff according to the ID of the customer service staff;
the audio screening module is used for selecting audio time length which is greater than a preset time length threshold value from the customer service audio to obtain candidate customer service audio;
the voice analysis module is used for extracting voice from the candidate customer service audio and performing text recognition on the voice to obtain a voice text of the candidate customer service audio; the voice frequency and the volume of the candidate customer service audio are detected to obtain the voice frequency and the volume of the candidate customer service audio;
the content extraction module is used for judging whether the candidate customer service audio comprises complaint content or not by adopting a preset complaint content judgment condition according to the voice text, the sound frequency and the volume of the candidate customer service audio, and storing the corresponding relation between the complaint content and the ID of the customer service staff in a complaint database if the candidate customer service audio comprises the complaint content; and judging whether the candidate customer service audio comprises the showing content or not by adopting a preset showing content judgment condition according to the voice text, the sound frequency and the volume of the candidate customer service audio, and if so, storing the corresponding relation between the showing content and the ID of the customer service personnel in a showing database.
10. The audio processing system for voice customer service evaluation according to claim 9, wherein the content extraction module adopts a preset complaint content judgment condition to judge whether the candidate customer service audio includes a complaint content by the following steps:
the content extraction module judges whether the candidate customer service audio comprises a preset complaint keyword or not;
if yes, the content extraction module judges whether the sound frequency of the part, corresponding to the complaint keyword, of the candidate customer service audio is greater than a first frequency threshold and the sound volume is greater than a first sound volume threshold, if yes, the candidate customer service audio comprises the complaint content, and if not, the candidate customer service audio does not comprise the complaint content;
if not, then the candidate customer service audio does not include complaint content;
the content extraction module adopts the following steps to adopt a preset suggestive content judgment condition to judge whether the candidate customer service audio comprises suggestive content:
the content extraction module judges whether the candidate customer service audio comprises a preset raise keyword or not;
if yes, the content extraction module judges whether the sound frequency of the part, corresponding to the showing keyword, in the candidate customer service audio is larger than a second frequency threshold, and the volume is larger than a second volume threshold, if yes, the candidate customer service audio comprises the showing content, otherwise, the candidate customer service audio does not comprise the showing content;
and if not, the candidate customer service audio does not comprise the showing content.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011106805.6A CN112235468A (en) | 2020-10-16 | 2020-10-16 | Audio processing method and system for voice customer service evaluation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011106805.6A CN112235468A (en) | 2020-10-16 | 2020-10-16 | Audio processing method and system for voice customer service evaluation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112235468A true CN112235468A (en) | 2021-01-15 |
Family
ID=74118032
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011106805.6A Withdrawn CN112235468A (en) | 2020-10-16 | 2020-10-16 | Audio processing method and system for voice customer service evaluation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112235468A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112911072A (en) * | 2021-01-28 | 2021-06-04 | 携程旅游网络技术（上海）有限公司 | Call center volume identification method and device, electronic equipment and storage medium |
| CN113392259A (en) * | 2021-05-14 | 2021-09-14 | 深圳航天科技创新研究院 | Sound stimulation sample selection method, control device and storage medium |
| CN117153185A (en) * | 2023-10-31 | 2023-12-01 | 建信金融科技有限责任公司 | Call processing method, device, computer equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102625005A (en) * | 2012-03-05 | 2012-08-01 | 广东天波信息技术股份有限公司 | Call center system with function of real-timely monitoring service quality and implement method of call center system |
| CN106024015A (en) * | 2016-06-14 | 2016-10-12 | 上海航动科技有限公司 | Call center agent monitoring method and system |
| CN106548788A (en) * | 2015-09-23 | 2017-03-29 | 中国移动通信集团山东有限公司 | A kind of intelligent emotion determines method and system |
| CN208656882U (en) * | 2018-09-17 | 2019-03-26 | 重庆先特服务外包产业有限公司 | Call center's traffic administration system |
| CN109785862A (en) * | 2019-01-21 | 2019-05-21 | 深圳壹账通智能科技有限公司 | Customer service quality evaluating method, device, electronic equipment and storage medium |
-
2020
- 2020-10-16 CN CN202011106805.6A patent/CN112235468A/en not_active Withdrawn
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102625005A (en) * | 2012-03-05 | 2012-08-01 | 广东天波信息技术股份有限公司 | Call center system with function of real-timely monitoring service quality and implement method of call center system |
| CN106548788A (en) * | 2015-09-23 | 2017-03-29 | 中国移动通信集团山东有限公司 | A kind of intelligent emotion determines method and system |
| CN106024015A (en) * | 2016-06-14 | 2016-10-12 | 上海航动科技有限公司 | Call center agent monitoring method and system |
| CN208656882U (en) * | 2018-09-17 | 2019-03-26 | 重庆先特服务外包产业有限公司 | Call center's traffic administration system |
| CN109785862A (en) * | 2019-01-21 | 2019-05-21 | 深圳壹账通智能科技有限公司 | Customer service quality evaluating method, device, electronic equipment and storage medium |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112911072A (en) * | 2021-01-28 | 2021-06-04 | 携程旅游网络技术（上海）有限公司 | Call center volume identification method and device, electronic equipment and storage medium |
| CN113392259A (en) * | 2021-05-14 | 2021-09-14 | 深圳航天科技创新研究院 | Sound stimulation sample selection method, control device and storage medium |
| CN117153185A (en) * | 2023-10-31 | 2023-12-01 | 建信金融科技有限责任公司 | Call processing method, device, computer equipment and storage medium |
| CN117153185B (en) * | 2023-10-31 | 2024-01-30 | 建信金融科技有限责任公司 | Call processing method, device, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112235468A (en) | Audio processing method and system for voice customer service evaluation | |
| US9530139B2 (en) | Evaluation of voice communications | |
| US8219404B2 (en) | Method and apparatus for recognizing a speaker in lawful interception systems | |
| US20100332287A1 (en) | System and method for real-time prediction of customer satisfaction | |
| CN107154257A (en) | Customer service quality evaluating method and system based on customer voice emotion | |
| CN107293309B (en) | Method for improving public opinion monitoring efficiency based on client emotion analysis | |
| CN110890096A (en) | Intelligent voice system and method based on voice analysis | |
| CN112261230B (en) | Express call answering rate assessment method, device, equipment, system and storage medium | |
| CN111597818B (en) | Call quality inspection method, device, computer equipment and computer readable storage medium | |
| TW201935370A (en) | System and method for evaluating customer service quality from text content | |
| CN110929011A (en) | Conversation analysis method, device and equipment | |
| CN110942326A (en) | User satisfaction prediction method and device and server | |
| US20230252486A1 (en) | Systems and methods for detecting complaint interactions | |
| CN111445928A (en) | Voice quality inspection method, device, equipment and storage medium | |
| CN112860868A (en) | Customer service telephone analysis method, system, equipment and storage medium | |
| CA3182191A1 (en) | Voice quality inspection method and device, computer equipment and storage medium | |
| CN108648768A (en) | A kind of consulting recommendation method and its management system | |
| CN114049895B (en) | ASR-based voice quality inspection analysis method and system | |
| CN112019692A (en) | Telephone traffic record storage method, system and computer readable medium | |
| CN113297365A (en) | User intention determination method, device, equipment and storage medium | |
| CN116828109A (en) | Intelligent evaluation method and system for telephone customer service quality | |
| CN111916110A (en) | Voice quality inspection method and device | |
| CN116013316A (en) | Content analysis method, device, equipment and medium for teleconference | |
| CN114356982A (en) | Marketing compliance checking method and device, computer equipment and storage medium | |
| CN113191711A (en) | Express delivery sending strategy determining method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20210115 |