{"title":"Deploying usable speech enabled IVR systems for mass use","authors":"Chitralekha Bhat, B. Mithun, Vikram Saxena, Vrushali Kulkarni, Sunil Kumar Kopparapu","doi":"10.1109/ICHCI-IEEE.2013.6887794","DOIUrl":null,"url":null,"abstract":"Interactive Voice Response (IVR) technology makes customer service 24 × 7 and cost effective and has been used by most customer facing enterprises. While effective, IVRs requires inputs to be keyed in using a touch tone phone, constraining the type of information that can be input. For this reason customers in general choose to speak with a human agent directly, negating the cost effectiveness of deploying an IVR. The need to speak to a human agent is more profound in a multi linguistic country like India, where in addition the agent has to necessarily speak in the language of the customer, making it unscalable. Speech technology based IVR solutions not only have all the benefits of a conventional IVR system but in addition address the scalability issue when configured to work in multiple languages. In order to enable a usable Speech enabled IVR into a complete and robust solution, several aspects need to be handled to sustain the core objective of serving the customer accurately. In this paper, we use, Speech Enabled Railway Enquiry System (SERES) which has been designed to provide railway information keeping in mind the Indian scenario, as a case study to identify issues that need to be addressed to enable a usable speech based IVR solution. Aspects looked at are a) Interaction design for customer experience b) field trial strategy c) System performance monitoring and evaluation and d) Infrastructure sizing to cater to the expected call volumes.","PeriodicalId":419263,"journal":{"name":"2013 International Conference on Human Computer Interactions (ICHCI)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2013-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"12","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2013 International Conference on Human Computer Interactions (ICHCI)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICHCI-IEEE.2013.6887794","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 12
Abstract
Interactive Voice Response (IVR) technology makes customer service 24 × 7 and cost effective and has been used by most customer facing enterprises. While effective, IVRs requires inputs to be keyed in using a touch tone phone, constraining the type of information that can be input. For this reason customers in general choose to speak with a human agent directly, negating the cost effectiveness of deploying an IVR. The need to speak to a human agent is more profound in a multi linguistic country like India, where in addition the agent has to necessarily speak in the language of the customer, making it unscalable. Speech technology based IVR solutions not only have all the benefits of a conventional IVR system but in addition address the scalability issue when configured to work in multiple languages. In order to enable a usable Speech enabled IVR into a complete and robust solution, several aspects need to be handled to sustain the core objective of serving the customer accurately. In this paper, we use, Speech Enabled Railway Enquiry System (SERES) which has been designed to provide railway information keeping in mind the Indian scenario, as a case study to identify issues that need to be addressed to enable a usable speech based IVR solution. Aspects looked at are a) Interaction design for customer experience b) field trial strategy c) System performance monitoring and evaluation and d) Infrastructure sizing to cater to the expected call volumes.