Domain specific voice intent classification with BLSTM

dc.contributor.advisorUthayasanker T
dc.contributor.authorHellarawa HMMJ
dc.date.accept2022
dc.date.accessioned2022
dc.date.available2022
dc.date.issued2022
dc.description.abstractWith the current global pandemic all countries around the globe are facing difficulties managing their healthcare services in a way that ensures the high availability of critical services while maintaining the safety of both the patient and the staff. According to Gartner’s top 10 strategic technology trends 2021 [1], it says “Rather than building a technology stack and then exploring the potential applications, organizations must consider the business and human context first.” where it highlights the need for human centric development while stating that it is the IT leaders that decides what combination of the trends to involve in driving the most innovation and strategy. A decade ago, simply having a website was enough to impress prospective customers and help them find their way to a service or information need and to establish a brand loyalty or identity. The growth of the technology is demanding more innovative strategies to adopted to every small to large industries that are at any stage of maturity of their roadmap to success. The increasing demands of the clients and the ability to keep a loyal customer base has highlighted the need of having a more natural way of handling a customer’s inquiry gives a competitive advantage for any business. The disappointment due to a customer getting added to a call waiting queues to reach a particular service is very critical and can even cause a loss of business opportunity. Understanding call intents can help a service provider to adapt the business engagement with the outside in a way that customers are positively satisfied which could in return increases the sales revenue. Not only that, but indirectly enables the ability for business to allocation agents or help-desk staff optimally thus avoid understaffing and overstaffing situation, which are indirect costs for any revenue-based figure. Automation is where the technology is used to automate tasks that once required humans. Here, the menu-based call center automations can be taken as a replacement to the legacy call center agent where the human tasks were replaced by automation. The concept of hyperautomation is where the businesses are rapidly adopting it’s revenue-based processes and IT process for automation. And the current state-of-the-art deals with lot of advanced technologies like Machine Learning (ML) and Artificial Intelligence (AI). Where AI and ML are used for extending the capabilities of automations.en_US
dc.identifier.accnoTH4933en_US
dc.identifier.citationHellarawa, H.M.M.J. (2022). Domain specific voice intent classification with BLSTM [Master's theses, University of Moratuwa]. Institutional Repository University of Moratuwa. http://dl.lib.uom.lk/handle/123/22390
dc.identifier.degreeMSc in Computer Science & Engineeringen_US
dc.identifier.departmentDepartment of Computer Science & Engineeringen_US
dc.identifier.facultyEngineeringen_US
dc.identifier.urihttp://dl.lib.uom.lk/handle/123/22390
dc.language.isoenen_US
dc.subjectVOICE INTENT CLASSIFICATIONen_US
dc.subjectHEALTHCAREen_US
dc.subjectBIDIRECTIONAL LONG SHORT-TERM MEMORY (BLSTM)en_US
dc.subjectSPEECH RECOGNITIONen_US
dc.subjectCOMPUTER SCIENCE & ENGINEERING - Dissertationen_US
dc.subjectCOMPUTER SCIENCE- Dissertationen_US
dc.titleDomain specific voice intent classification with BLSTMen_US
dc.typeThesis-Abstracten_US

Files

Original bundle

Now showing 1 - 3 of 3
Loading...
Thumbnail Image
Name:
TH4933-1.pdf
Size:
470.15 KB
Format:
Adobe Portable Document Format
Description:
Pre-Text
Loading...
Thumbnail Image
Name:
TH4933-2.pdf
Size:
146.95 KB
Format:
Adobe Portable Document Format
Description:
Post- Text
Loading...
Thumbnail Image
Name:
TH4933.pdf
Size:
2.2 MB
Format:
Adobe Portable Document Format
Description:
Full theses