[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-126648-en":3,"doc-seo-126648-105":31,"detail-sidebar-cat-0-en-105":92},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":28,"seo_description":14,"update_tm":29,"read_time":30},126648,549768064778,"Finn","https://ap-avatar.wpscdn.com/davatar_6f874abed73319feea01a86fa6f0fab8",6,"Technology","Call Audio Quality Determination and Root Cause Analysis Using Machine Learning","Accurate assessment and categorization of real-world audio quality during calls over VoLTE/VoNR is essential for a satisfactory user experience, yet existing techniques cannot reliably label call audio conditions such as mild choppiness, severe choppiness, or complete absence of audio. The disclosure presents a closed-loop machine-learning approach that clusters audio metrics, derives a root cause table from cluster patterns, and trains a classifier to detect unsatisfactory audio quality in ongoing calls. When quality is poor, the likely root cause is identified to enable solutions while the call is in progress.","Technical Disclosure Commons  \nDefensive Publications Series  \nMarch 2023  \nCall Audio Quality Determination and Root Cause Analysis Using Machine Learning  \nYu Liu  \nJack Chan  \nQin Zhang  \nXiantao Sun  \nFollow this and additional works at: [https://www.tdcommons.org/dpubs_series](https://www.tdcommons.org/dpubs_series)  \nRecommended Citation  \nLiu, Yu; Chan, Jack; Zhang, Qin; and Sun, Xiantao, \"Call Audio Quality Determination and Root Cause Analysis Using Machine Learning\", Technical Disclosure Commons,(March 30, 2023)  \n[https://www.tdcommons.org/dpubs_series/5772](https://www.tdcommons.org/dpubs_series/5772)  \nThis work is licensed under a Creative Commons Attribution 4.0 License.  \nThis Article is brought to you for free and open access by Technical Disclosure Commons. It has been accepted for inclusion in Defensive Publications Series by an authorized administrator of Technical Disclosure Commons.  \nCall Audio Quality Determination and Root Cause Analysis Using Machine Learning  \nABSTRACT  \nAccurate assessment and categorization of real-world audio quality in a call, e.g., a call  \nover VoLTE/VoNR, is essential to provide a satisfactory call experience. However, current  \ntechniques to determine call quality do not accurately categorize the audio quality. Also, there  \nare no techniques to determine the root cause of poor audio quality or to identify potential  \nsolutions. This disclosure describes the use of machine learning clustering techniques to cluster  \naudio metrics and using the obtained clusters to generate a root cause table. Further, a classifier  \nis trained to determine whether an ongoing call has unsatisfactory audio quality. The quality can  \nbe categorized and labeled, e.g., good, mildly choppy, severely choppy, and no audio. If the  \naudio quality is unsatisfactory, the likely root cause is identified using the root cause table to  \nidentify and apply solutions while the call is in progress. The described techniques are a closed  \nloop technique to identify solutions to audio quality problems in an audio call.  \nKEYWORDS  \n● Root cause analysis ● Clustering algorithm  \n● Root cause table ● Audio quality metrics  \n● Choppy audio ● Voice over LTE (VoLTE)  \n● Audio quality ● Voice over New Radio (VoNR)  \n● Voice quality  \nBACKGROUND  \nMobile users that communicate over Voice over LTE (VoLTE) or Voice over New Radio (VoNR) may experience instances of pauses, mute (no audio from other side), and/or choppy audio during call sessions even as the real-time voice call continues with a connected network.  \nPublished by Technical Disclosure Commons, 2023 2  \nThis is a suboptimal experience and occurs when the participating devices are experiencing poor quality conditions. Efficiently detecting real-world call performance and fixing the poor quality issues is necessary to improve the user experience. However, there are no current techniques to  \neither evaluate the real-time audio quality, to identify root cause of poor quality, or to identify  \nrecommended solutions.  \nA wireless modem can provide network traffic statistics such as real-time transport protocol (RTP) packet loss, delay, audio codec rates, network performance, etc. Poor call quality can be caused by many factors such as poor network connections, modem software issues,  \nhardware issues, etc. Some techniques (e.g., [1]) to determine reasons for poor audio quality  \nfocus on calculating R factors or mean opinion score (MOS) as performance metrics of call  \nquality. However, such techniques cannot accurately categorize the real world call quality and  \nconditions such as mild choppy, severe choppy, or no audio. There is no work to determine the  \nroot cause for poor quality audio. There are no closed-loop systems that can provide  \nrecommendations of actions to improve audio quality.  \nDESCRIPTION  \nThis disclosure describes the use of machine learning clustering techniques to cluster audio metrics and using the obtained clusters to generate a root ca","cbCaiqNzhGsXGcuX","https://ap.wps.com/l/cbCaiqNzhGsXGcuX","pdf",348736,4,1,7,"English","en",105,"# Background\n## Call quality issues in VoLTE/VoNR\n## Limitations of existing audio-quality techniques\n# Description\n## Clustering audio metrics with machine learning\n## Training a classifier for ongoing-call quality\n## Generating and using a root cause table\n# Example Method (Fig. 1)\n## Gathering call data with user permission\n## Obtaining relevant metrics\n## Training, evaluating, and tuning a clustering model\n## Cluster pattern analysis to generate root causes","[{\"question\":\"What problem does the disclosure address in call audio quality assessment?\",\"answer\":\"It addresses the inability of current techniques to accurately categorize real-world audio quality and the lack of methods to determine root causes of poor audio or identify solutions during the call.\"},{\"question\":\"How are audio quality categories such as good or choppy determined?\",\"answer\":\"A machine learning workflow clusters audio metrics and trains a classifier to label ongoing calls, including categories like good, mildly choppy, severely choppy, and no audio.\"},{\"question\":\"How does the system identify likely root causes of unsatisfactory audio quality?\",\"answer\":\"It generates a root cause table by performing cluster pattern analysis on clustered metrics, then uses that table to map an unsatisfactory call to the likely underlying cause while the call is still in progress.\"}]","Call Audio Quality Determination and Root Cause Analysis Using Machine Learning | PDF",1785934027,18,{"code":4,"msg":32,"data":33},"ok",{"site_id":25,"language":24,"slug":34,"title":13,"keywords":35,"description":14,"schema_data":36,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":29},"call-audio-quality-determination-and-root-cause-analysis-using-machine-learning","",{"@graph":37,"@context":86},[38,54,69],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,49,52],{"item":42,"name":43,"@type":44,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":46,"name":47,"@type":44,"position":48},"https://docshare.wps.com/document/","Document",2,{"item":50,"name":12,"@type":44,"position":51},"https://docshare.wps.com/document/technology/",3,{"item":53,"name":13,"@type":44,"position":20},"https://docshare.wps.com/document/call-audio-quality-determination-and-root-cause-analysis-using-machine-learning/126648/",{"url":53,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":24,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":42,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-28","2026-08-05",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What problem does the disclosure address in call audio quality assessment?","Question",{"text":76,"@type":77},"It addresses the inability of current techniques to accurately categorize real-world audio quality and the lack of methods to determine root causes of poor audio or identify solutions during the call.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How are audio quality categories such as good or choppy determined?",{"text":81,"@type":77},"A machine learning workflow clusters audio metrics and trains a classifier to label ongoing calls, including categories like good, mildly choppy, severely choppy, and no audio.",{"name":83,"@type":74,"acceptedAnswer":84},"How does the system identify likely root causes of unsatisfactory audio quality?",{"text":85,"@type":77},"It generates a root cause table by performing cluster pattern analysis on clustered metrics, then uses that table to map an unsatisfactory call to the likely underlying cause while the call is still in progress.","https://schema.org",{"og:url":53,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":53},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":93},[94,98,102,106,111,114,118,123,128,131,135],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":48,"doc_module":4,"doc_module_name":47,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":20,"doc_module":4,"doc_module_name":47,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":47,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":11,"doc_module":4,"doc_module_name":47,"category_name":12,"show_sort_weight":112,"slug":113},50,"technology",{"id":22,"doc_module":4,"doc_module_name":47,"category_name":115,"show_sort_weight":116,"slug":117},"Healthcare",40,"healthcare",{"id":119,"doc_module":4,"doc_module_name":47,"category_name":120,"show_sort_weight":121,"slug":122},8,"Research & Report",30,"research-report",{"id":124,"doc_module":4,"doc_module_name":47,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":47,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":47,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":47,"category_name":137,"show_sort_weight":107,"slug":138},19,"General","general"]