speech analytics

Speech analytics provides a Complete analysis of recorded phone conversations between a company and its customers.{{cite web|url=http://searchcrm.techtarget.com/report/Top-five-benefits-of-speech-analytics-for-the-call-center|title=Top five benefits of speech analytics for the call center|publisher=TechTarget}} It provides advanced functionality and valuable intelligence from customer calls. This information can be used to discover information relating to strategy, product, process, operational issues and contact center agent performance.{{cite web|url=http://www.genesys.com/platform-services/workforce-optimization/speech-text-analytics|title=Speech & Text Analytics|publisher=Genesys}} In addition, speech analytics can automatically identify areas in which contact center agents may need additional training or coaching,{{cite web|url=https://www.xdroid.com/why-reduction-of-silence-periods-is-a-goal-and-reduction-of-the-average-handling-time-is-not|title=Real Time Voice Analytics|publisher=Xdroid}} and can automatically monitor the customer service provided on calls.{{cite web|url=http://www.icmi.com/Resources/Learning-and-Development/2015/04/Do-Speech-Analytics-Tools-Change-Agent-Behavior|title=Do Speech Analytics Tools Change Agent Behavior?|publisher=ICMI}}

The process can isolate the words and phrases used most frequently within a given time period, as well as indicate whether usage is trending up or down. This information is useful for supervisors, analysts, and others in an organization to spot changes in consumer behavior and take action to reduce call volumes—and increase customer satisfaction. It allows insight into a customer's thought process, which in turn creates an opportunity for companies to make adjustments.{{cite web|url=https://www.entrepreneur.com/article/241112|title=Reverse a Pattern of Poor Sales With Speech Analytics|publisher=Entrepreneur}}

Speech analytics applications can spot spoken keywords or phrases, either as real-time alerts on live audio or as a post-processing step on recorded speech. This technique is also known as audio mining. Other uses include categorization of speech in the contact center environment to identify calls from unsatisfied customers.{{cite web|url=http://www.destinationcrm.com/Articles/Editorial/Magazine-Features/The-Age-of-Speech-Analytics-Is-Close-at-Hand-106676.aspx|title=The Age of Speech Analytics Is Close at Hand|last=|first=|date=|website=|publisher=Destination CRM|access-date=30 September 2016}}

Measures such as Precision and recall, commonly used in the field of Information retrieval, are typical ways of quantifying the response of a speech analytics search system.C. D. Manning, P. Raghavan, and H. Schütze, Introduction to Information Retrieval, Chapter 8. Precision measures the proportion of search results that are relevant to the query. Recall measures the proportion of the total number of relevant items that were returned by the search results. Where a standardised test set has been used, measures such as precision and recall can be used to directly compare the search performance of different speech analytics systems.

Making a meaningful comparison of the accuracy of different speech analytics systems can be difficult. The output of LVCSR systems can be scored against reference word-level transcriptions to produce a value for the word error rate (WER), but because phonetic systems use phones as the basic recognition unit, rather than words, comparisons using this measure cannot be made. When speech analytics systems are used to search for spoken words or phrases, what matters to the user is the accuracy of the search results that are returned. Because the impact of individual recognition errors on these search results can vary greatly, measures such as word error rate are not always helpful in determining overall search accuracy from the user perspective.

According to the US Government Accountability Office,{{cite web|url=http://www.gao.gov/new.items/d03273g.pdf|title=Assessing the Reliability of Computer-Processed Data|publisher=United States General Accounting Office}} “data reliability refers to the accuracy and completeness of computer-processed data, given the uses they are intended for.” In the realm of Speech Recognition and Analytics, “completeness” is measured by the “detection rate”, and usually as accuracy goes up, the detection rate goes down.{{cite web |url=https://knowledgespace.com.au/what-does-speech-analytics-software-actually-do/ |url-status=dead |archive-url=https://web.archive.org/web/20180123131643/https://knowledgespace.com.au/what-does-speech-analytics-software-actually-do/ |archive-date=2018-01-23 |title=What Does Speech Analytics Software Actually Do? - KnowledgeSpace}}

Speech analytics vendors use the "engine" of a 3rd party and others develop proprietary engines. The technology mainly uses three approaches. The phonetic approach is the fastest for processing, mostly because the size of the grammar is very small, with a phoneme as the basic recognition unit. There are only few tens of unique phonemes in most languages, and the output of this recognition is a stream (text) of phonemes, which can then be searched. Large-vocabulary continuous speech recognition (LVCSR, more commonly known as speech-to-text, full transcription or ASR - automatic speech recognition) uses a set of words (bi-grams, tri-grams etc.) as the basic unit. This approach requires hundreds of thousands of words to match the audio against. It can surface new business issues, the queries are much faster, and the accuracy is higher than the phonetic approach.{{cite web|url=http://www.callminer.com/wp-content/whitepapers/The-Right-Technology-for-Speech-Analytics.pdf|title=The Right Technology for your Speech Analytics Project|publisher=CallMiner|access-date=30 September 2016}}

Extended speech emotion recognition and prediction is based on three main classifiers: kNN, C4.5 and SVM RBF Kernel. This set achieves better performance than each basic classifier taken separately. It is compared with two other sets of classifiers: one-against-all (OAA) multiclass SVM with Hybrid kernels and the set of classifiers which consists of the following two basic classifiers: C5.0 and Neural Network. The proposed variant achieves better performance than the other two sets of classifiers.{{cite journal|year=2014|title=Extended speech emotion recognition and prediction|url=http://ntv.ifmo.ru/en/article/11200/raspoznavanie_i_prognozirovanie_dlitelnyh__emociy_v_rechi_(na_angl._yazyke).htm|journal=Scientific and Technical Journal of Information Technologies, Mechanics and Optics|volume=14|issue=6|page=137|author=S.E. Khoruzhnikov|display-authors=etal}}

Market research indicates that speech analytics is projected to become a billion dollar industry by 2020 with North America having the largest market share.{{cite web|url=http://www.prnewswire.com/news-releases/speech-analytics-market-worth-160-billion-usd-by-2020-575725271.html|title=Speech Analytics Market Worth 1.60 Billion USD by 2020|publisher=PR Newswire}} The growth rate is attributed to rising requirements for compliance and risk management as well as an increase in industry competition through market intelligence.{{cite web|url=http://www.menafn.com/1094850229/Speech-Analytics-Industry-Market-Share-Size-Growth--Forecast-2025|title=Speech Analytics Industry Market Share, Size, Growth & Forecast 2025|publisher=MENAFN}} The telecommunications, IT and outsourcing segments of the industry are considered to hold the largest market share with expected growth from the travel and hospitality segments.

• Customer intelligence
• Customer dynamics
• Speech recognition

{{Reflist}}

{{Computer audition}}

{{DEFAULTSORT:Speech analytics}}

Category:Speech recognition

Category:Customer relationship management