Crafting ASR and Conversational Models for an Agriculture Chatbot

International Conference on Computational Intelligence and Intelligent Systems Pub Date : 1900-01-01 DOI:10.1145/3507623.3507634

Abbott Po Shun Chen, C. Liu

{"title":"Crafting ASR and Conversational Models for an Agriculture Chatbot","authors":"Abbott Po Shun Chen, C. Liu","doi":"10.1145/3507623.3507634","DOIUrl":null,"url":null,"abstract":"In recent years, artificial intelligence chatbots have attracted more and more attention. The stability and accuracy of automatic speech recognition (ASR) have been improved, making voice more critical in the transaction process and voice consultation of e-commerce purchases. ASR matches the learning model based on contextual cues. Eliminating unnecessary text plays an important role. We use the LSTM model and change it to contextualized custom text. In addition, to use our robot for testing, we propose a multi-task model that can jointly perform content re-scoring and has excellent responsiveness in the text of the input entity. Therefore, this article recommends using ASR technology to interpret and predict the answers to the LSTM model, allowing users to obtain expected results from actual measurements and understand which aspects are suitable for predicting specific emotions tested on this group. This article discusses chatbots and the design techniques used in platform translation, early and modern chatbots combined with ASR and artificial intelligence technology.","PeriodicalId":159712,"journal":{"name":"International Conference on Computational Intelligence and Intelligent Systems","volume":"48 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Conference on Computational Intelligence and Intelligent Systems","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3507623.3507634","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 1

Abstract

In recent years, artificial intelligence chatbots have attracted more and more attention. The stability and accuracy of automatic speech recognition (ASR) have been improved, making voice more critical in the transaction process and voice consultation of e-commerce purchases. ASR matches the learning model based on contextual cues. Eliminating unnecessary text plays an important role. We use the LSTM model and change it to contextualized custom text. In addition, to use our robot for testing, we propose a multi-task model that can jointly perform content re-scoring and has excellent responsiveness in the text of the input entity. Therefore, this article recommends using ASR technology to interpret and predict the answers to the LSTM model, allowing users to obtain expected results from actual measurements and understand which aspects are suitable for predicting specific emotions tested on this group. This article discusses chatbots and the design techniques used in platform translation, early and modern chatbots combined with ASR and artificial intelligence technology.

查看原文