{"title":"RED:基于智能边缘的扬声器系统与环境传感技术","authors":"Tanvi K Jois, M. V. Bharadwaj, A. Mukhopadhyay","doi":"10.1109/ICDSIS55133.2022.9915823","DOIUrl":null,"url":null,"abstract":"Smart speakers have become widely common. A smart speaker allows people to do a wide range of tasks, including presenting information such as the time, date, and several other things. It can also stream songs on its own. Smart speakers that can carry out tasks are known as virtual assistants. In this work, a smart speaker RED is demonstrated. RED is particularly designed for audio sensing activities. Mel’s spectrograms are obtained and subjected to rudimentary neural networks to achieve automatic speech recognition in RED. RED can adjust to ambient noise, communicate with the user, and function as a virtual assistant by cracking jokes, playing/pausing music, and altering volume on command. Users can dynamically assign wake words of their desire in RED. RED was able to reach an overall accuracy of 91%.","PeriodicalId":178360,"journal":{"name":"2022 IEEE International Conference on Data Science and Information System (ICDSIS)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"RED: An Intelligent Edge based Speaker System with Ambient Sensing Technology\",\"authors\":\"Tanvi K Jois, M. V. Bharadwaj, A. Mukhopadhyay\",\"doi\":\"10.1109/ICDSIS55133.2022.9915823\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Smart speakers have become widely common. A smart speaker allows people to do a wide range of tasks, including presenting information such as the time, date, and several other things. It can also stream songs on its own. Smart speakers that can carry out tasks are known as virtual assistants. In this work, a smart speaker RED is demonstrated. RED is particularly designed for audio sensing activities. Mel’s spectrograms are obtained and subjected to rudimentary neural networks to achieve automatic speech recognition in RED. RED can adjust to ambient noise, communicate with the user, and function as a virtual assistant by cracking jokes, playing/pausing music, and altering volume on command. Users can dynamically assign wake words of their desire in RED. RED was able to reach an overall accuracy of 91%.\",\"PeriodicalId\":178360,\"journal\":{\"name\":\"2022 IEEE International Conference on Data Science and Information System (ICDSIS)\",\"volume\":\"25 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2022-07-29\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2022 IEEE International Conference on Data Science and Information System (ICDSIS)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICDSIS55133.2022.9915823\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 IEEE International Conference on Data Science and Information System (ICDSIS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICDSIS55133.2022.9915823","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
RED: An Intelligent Edge based Speaker System with Ambient Sensing Technology
Smart speakers have become widely common. A smart speaker allows people to do a wide range of tasks, including presenting information such as the time, date, and several other things. It can also stream songs on its own. Smart speakers that can carry out tasks are known as virtual assistants. In this work, a smart speaker RED is demonstrated. RED is particularly designed for audio sensing activities. Mel’s spectrograms are obtained and subjected to rudimentary neural networks to achieve automatic speech recognition in RED. RED can adjust to ambient noise, communicate with the user, and function as a virtual assistant by cracking jokes, playing/pausing music, and altering volume on command. Users can dynamically assign wake words of their desire in RED. RED was able to reach an overall accuracy of 91%.